FIELD OF DISCLOSURE

[0001] The present disclosure relates to the field of image processing. More particularly, the present disclosure relates to system and method for detecting a vehicle during day and night time.

BACKGROUND OF THE DISCLOSURE

[0002] A robust and reliable vehicle detection system is one of the key elements in vehicle automation, which makes accurate and precise vehicle detection a subject of prime importance. Due to change in visibility of vehicles on the road owing to various conditions such as weather, glare, pollution or inherent human weakness many detection systems and devices utilizing different techniques have been developed. Many existing techniques operable to detect vehicles or other objects in vicinity of a host vehicle are based on sensors such as Radio Detection and Ranging (RADAR), Light Detection and Ranging (LIDAR), ultrasound, camera and the like incorporated in the host vehicle.

[0003] Certain exiting techniques are based on radar data which is useful for detecting and measuring distance of target vehicle, however, these techniques suffer from false detection, which leads to lower efficiency in detection in terms of accuracy and places reliability of such system under scrutiny. Certain other existing techniques utilize statistical methodologies to statistically analyse data obtained from various sensors implemented in the host vehicle to estimate whether a vehicle is detected by the host vehicle. Such techniques typically use statistical methodologies to determine likelihood or probability that a detected object is a vehicle. However, such techniques also lead to generation of false positive detections and thereby reduce efficiency. Further, certain exiting techniques are based on computer vision based approaches. As compared to other approaches, vision-based systems are gaining significant importance due to their lower cost and advantages as compared to other sensors.

[0004] However, vision based techniques may fail to detect the presence of a vehicle if visibility of the vehicle is poor, for example, a computer vision based detection system may not detect a black car during night time. Also, as different camera sensors are used depending upon day or night conditions, undesired complications are created in the system. Therefore, the major challenge of the existing approaches lies while dealing with complex scenario and different algorithms for different ambient light conditions e.g. day and night. Furthermore, the challenges also increase while detecting correct location e.g. bottom point of the vehicle for stabilized Automatic Emergency Braking (AEB) and Automatic Cruise Control (ACC). Also, due to unpredictable motion of vehicles, trajectory prediction of target vehicles becomes difficult. As existing techniques face above-mentioned and other disadvantages, they may hinder in development of efficient AEB and/or ACC systems that are other key elements of vehicle automation.

[0005] There is therefore need in the art to develop a system and method for detection of vehicles especially during day and night time that overcomes the above-mentioned and other limitations of the existing solutions and utilize techniques, which are robust, accurate, fast, efficient and simple.

OBJECTS OF THE PRESENT DISCLOSURE

[0006] Some of the objects of the present disclosure, which at least one embodiment herein satisfies are as listed herein below.

[0007] It is an object of the present disclosure to provide a system and method for detecting a vehicle during day and night time.

[0008] It is an object of the present disclosure to provide a system and method for detecting a vehicle that provides appropriate warnings on detection.

[0009] It is yet another object of the present disclosure to provide a system and method for detecting a vehicle that tracks unpredictable motion of the vehicle.

[00010] It is yet another object of the present disclosure to provide a system and method for detecting distance of a vehicle for smoother Automatic Emergency Braking (AEB) and Automatic Cruise Control (ACC) in complex scenarios.

[00011] It is yet another object of the present disclosure to provide a system and method for detecting vehicles with different back portions.

[00012] It is still another object of the present disclosure to provide a robust, economic and simple system and method that accurately detects a vehicle. SUMMARY

[00013] This summary is provided to introduce simplified concepts of a system and method for vehicle detection, which are further described below in the Detailed Description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended for use in determining/limiting the scope of the claimed subject matter.

[00014] The present disclosure relates to the field of image processing. More particularly, the present disclosure relates to system and method for detecting a vehicle during day and night time.

[00015] An aspect of the present disclosure provides a system implemented in a host vehicle for detecting a vehicle during day and night time, the system comprises: an input unit comprising an image sensor for imaging field of view of the host vehicle; a processing unit comprising a processor coupled with a memory, the memory storing instructions executable by the processor to: receive one or more images from the input unit and define a Region of Interest (ROI) for a set of images selected from the one or more images, wherein the ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the host vehicle, and wherein a local histogram equalization-based contrast enhancement technique is performed in the ROI to make the output independent of illumination variation; determine a plurality of scanning windows in the ROI of each image based on categorization of the vehicle of the vehicle into any of a commercial vehicle or a passenger vehicle, wherein size of each scanning window is computed based on presumed height and width of the vehicle in the image; extract Integral Histogram of oriented gradients (IHOG) features from each of the scanning window based on respective feature descriptor of a pre-defined dimension; and detect the vehicle based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers, which is selected based on the categorization of the vehicle into any of the commercial vehicle or the passenger vehicle and performs the detection based on ambient light conditions, wherein at least one classifier of the two or more classifiers is trained separately for detecting the vehicle in each region defined in the field of view of the host vehicle, and wherein in response to the detection, distance of the detected vehicle from the host vehicle is determined. [00016] In an embodiment, the region in the field of view of the host vehicle is any of a far region, a middle region or a near region and wherein, the ROI is selected from High Definition (HD) resolution image for the far region, Video Graphics Array (VGA) resolution image for the middle region, and Quarter VGA (QVGA) resolution image for the near region.

[00017] In an embodiment, the processor further performs Non-maximal suppression (NMS) based on the categorization of the vehicle in order to provide an output of a single bounding box around the vehicle.

[00018] In an embodiment, the ambient light conditions are detected using state of head light switch of the host vehicle.

[00019] In an embodiment, the at least one classifier of the two or more classifier is trained using a training image dataset comprising training images, which are segregated based on resolution of each training image and the region defined in the field of view of the host vehicle.

[00020] In an embodiment, training of the at least one classifier is performed using the training image dataset by: cropping one or more training images of each of the training image dataset; resizing each of the cropped one or more images to size of a nearest scanning window; removing undesirable structures from a classifier database of the at least one classifier; collecting false positives from the at least one classifier; and adding the false positives to the classifier database of a succeeding classifier.

[00021] In an embodiment, in response to detection of ambient light conditions pertaining to night, false positives from output of the cascade of two or more classifiers are removed using V channel extraction.

[00022] In an embodiment, position of the detected vehicle is determined based on width of the vehicle.

[00023] In an embodiment, the processor tracks the detected vehicle by using a tracker selected based on categorization of the vehicle, wherein the tracker determines a state by analysing detection of the vehicle in consecutive images of the set of images, the state being determined from any of an idle state, a pre-track state, a tracking state or a cancel state, and wherein the tracker provides history-based motion prediction to track motion of the vehicle in case detection of the vehicle is missed. [00024] In an embodiment, the image sensor is a single Red Clear Clear Clear (RCCC) camera sensor.

[00025] Another aspect of the present disclosure provides a method for detecting a vehicle during day and night time, carried out according to instructions stored in a computer implemented in a host vehicle, comprising: receiving one or more images from an image sensor that images field of view of the host vehicle and defining a Region of Interest (ROI) for a set of images selected from the one or more images, wherein the ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the vehicle driver, and wherein a local histogram equalization-based contrast enhancement technique is performed in the ROI to make the output independent of illumination variation; determining a plurality of scanning windows in the ROI of each image based on categorization of the vehicle into any of a commercial vehicle or a passenger vehicle, wherein size of each scanning window is computed based on presumed height and width of the vehicle in the image; extracting Integral Histogram of oriented gradients (IHOG) features from each of the scanning window based on respective feature descriptor of a pre-defined dimension; and detecting the vehicle, based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers, which is selected based on the categorization of the vehicle into any of the commercial vehicle or the passenger vehicle, wherein at least one classifier of the two or more classifiers is trained separately for detecting the vehicle in each region defined in the field of view of the host vehicle, and wherein in response to the detection, distance of the detected vehicle from the host vehicle is determined.

[00026] Various objects, features, aspects and advantages of the present disclosure will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which like numerals represent like features.

[00027] Within the scope of this application it is expressly envisaged that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and/or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. Features described in connection with one embodiment are applicable to all embodiments, unless such features are incompatible. BRIEF DESCRIPTION OF DRAWINGS

[00028] The accompanying drawings are included to provide a further understanding of the present disclosure, and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure. The diagrams are for illustration only, which thus is not a limitation of the present disclosure, and wherein:

[00029] FIGs. 1A-B illustrate architecture of a vehicle detection system to illustrate its overall working in accordance with an embodiment of the present disclosure.

[00030] FIG. 2 illustrates exemplary modules of a processing unit in accordance with an embodiment of the present disclosure.

[00031] FIGs. 3A-F illustrate exemplary implementations of initialization and scanning window selection module in accordance with an exemplary embodiment of the present disclosure.

[00032] FIG. 4 illustrates an exemplary block diagram of feature extraction module in accordance with an embodiment of the present disclosure.

[00033] FIGs. 5A-C illustrate exemplary implementations of vehicle detection module in accordance with an exemplary embodiment of the present disclosure.

[00034] FIG. 6 illustrates an exemplary output from Non-maximal suppression (NMS) module in accordance with an embodiment of the present disclosure.

[00035] FIGs.7A-D illustrate exemplary implementations of false positive removal module in accordance with an exemplary embodiment of the present disclosure.

[00036] FIG. 8 illustrates an exemplary implementation of tracking module in accordance with an exemplary embodiment of the present disclosure.

[00037] FIG. 9 illustrates an exemplary block diagram for width based distance correction in accordance with an exemplary embodiment of the present disclosure.

[00038] FIG. 10 is a flow diagram illustrating a method for detection of a vehicle in accordance with an embodiment of the present disclosure.

[00039] FIGs. 11A-C illustrate exemplary testing results obtained in accordance with an embodiment of the present disclosure. DETAILED DESCRIPTION

[00040] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure as defined by the appended claims.

[00041] In the following description, numerous specific details are set forth in order to provide a thorough understanding of embodiments of the present invention. It will be apparent to one skilled in the art that embodiments of the present invention may be practiced without some of these specific details.

[00042] Embodiments of the present invention include various steps, which will be described below. The steps may be performed by hardware components or may be embodied in machine-executable instructions, which may be used to cause a general-purpose or special- purpose processor programmed with the instructions to perform the steps. Alternatively, steps may be performed by a combination of hardware, software, and firmware and/or by human operators.

[00043] Various methods described herein may be practiced by combining one or more machine-readable storage media containing the code according to the present invention with appropriate standard computer hardware to execute the code contained therein. An apparatus for practicing various embodiments of the present invention may involve one or more computers (or one or more processors within a single computer) and storage systems containing or having network access to computer program(s) coded in accordance with various methods described herein, and the method steps of the invention could be accomplished by modules, routines, subroutines, or subparts of a computer program product.

[00044] If the specification states a component or feature“may”,“can”,“could”, or “might” be included or have a characteristic, that particular component or feature is not required to be included or have the characteristic.

[00045] As used in the description herein and throughout the claims that follow, the meaning of“a,”“an,” and“the” includes plural reference unless the context clearly dictates otherwise. Also, as used in the description herein, the meaning of“in” includes“in” and“on” unless the context clearly dictates otherwise.

[00046] Exemplary embodiments will now be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments are shown. These exemplary embodiments are provided only for illustrative purposes and so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those of ordinary skill in the art. The invention disclosed may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Various modifications will be readily apparent to persons skilled in the art. The general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the invention. Moreover, all statements herein reciting embodiments of the invention, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future (i.e., any elements developed that perform the same function, regardless of structure). Also, the terminology and phraseology used is for the purpose of describing exemplary embodiments and should not be considered limiting. Thus, the present invention is to be accorded the widest scope encompassing numerous alternatives, modifications and equivalents consistent with the principles and features disclosed. For purpose of clarity, details relating to technical material that is known in the technical fields related to the invention have not been described in detail so as not to unnecessarily obscure the present invention.

[00047] Thus, for example, it will be appreciated by those of ordinary skill in the art that the diagrams, schematics, illustrations, and the like represent conceptual views or processes illustrating systems and methods embodying this invention. The functions of the various elements shown in the figures may be provided through the use of dedicated hardware as well as hardware capable of executing associated software. Similarly, any switches shown in the figures are conceptual only. Their function may be carried out through the operation of program logic, through dedicated logic, through the interaction of program control and dedicated logic, or even manually, the particular technique being selectable by the entity implementing this invention. Those of ordinary skill in the art further understand that the exemplary hardware, software, processes, methods, and/or operating systems described herein are for illustrative purposes and, thus, are not intended to be limited to any particular named element.

[00048] Embodiments of the present invention may be provided as a computer program product, which may include a machine-readable storage medium tangibly embodying thereon instructions, which may be used to program a computer (or other electronic devices) to perform a process. The term“machine-readable storage medium” or“computer-readable storage medium” includes, but is not limited to, fixed (hard) drives, magnetic tape, floppy diskettes, optical disks, compact disc read-only memories (CD-ROMs), and magneto-optical disks, semiconductor memories, such as ROMs, PROMs, random access memories (RAMs), programmable read-only memories (PROMs), erasable PROMs (EPROMs), electrically erasable PROMs (EEPROMs), flash memory, magnetic or optical cards, or other type of media/machine-readable medium suitable for storing electronic instructions (e.g., computer programming code, such as software or firmware). A machine-readable medium may include a non-transitory medium in which data may be stored and that does not include carrier waves and/or transitory electronic signals propagating wirelessly or over wired connections. Examples of a non-transitory medium may include, but are not limited to, a magnetic disk or tape, optical storage media such as compact disk (CD) or digital versatile disk (DVD), flash memory, memory or memory devices. A computer-program product may include code and/or machine-executable instructions that may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a class, or any combination of instructions, data structures, or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing and/or receiving information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, etc.

[00049] Furthermore, embodiments may be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments to perform the necessary tasks (e.g., a computer-program product) may be stored in a machine-readable medium. A processor(s) may perform the necessary tasks. [00050] Systems depicted in some of the figures may be provided in various configurations. In some embodiments, the systems may be configured as a distributed system where one or more components of the system are distributed across one or more networks in a cloud computing system.

[00051] Each of the appended claims defines a separate invention, which for infringement purposes is recognized as including equivalents to the various elements or limitations specified in the claims. Depending on the context, all references below to the "invention" may in some cases refer to certain specific embodiments only. In other cases it will be recognized that references to the "invention" will refer to subject matter recited in one or more, but not necessarily all, of the claims.

[00052] All methods described herein may be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context The use of any and all examples, or exemplary language (e.g.,“such as”) provided with respect to certain embodiments herein is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention otherwise claimed. No language in the specification should be construed as indicating any non-claimed element essential to the practice of the invention.

[00053] Various terms as used herein are shown below. To the extent a term used in a claim is not defined below, it should be given the broadest definition persons in the pertinent art have given that term as reflected in printed publications and issued patents at the time of filing.

[00054] This summary is provided to introduce simplified concepts of a system and method for vehicle detection, which are further described below in the Detailed Description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended for use in determining/limiting the scope of the claimed subject matter.

[00055] The present disclosure relates to the field of image processing. More particularly, the present disclosure relates to system and method for detecting a vehicle during day and night time.

[00056] An aspect of the present disclosure provides a system implemented in a host vehicle for detecting a vehicle during day and night time, the system comprises: an input unit comprising an image sensor for imaging field of view of the host vehicle; a processing unit comprising a processor coupled with a memory, the memory storing instructions executable by the processor to: receive one or more images from the input unit and define a Region of Interest (ROI) for a set of images selected from the one or more images, wherein the ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the host vehicle, and wherein a local histogram equalization-based contrast enhancement technique is performed in the ROI to make the output independent of illumination variation; determine a plurality of scanning windows in the ROI of each image based on categorization of the vehicle of the vehicle into any of a commercial vehicle or a passenger vehicle, wherein size of each scanning window is computed based on presumed height and width of the vehicle in the image; extract Integral Histogram of oriented gradients (IHOG) features from each of the scanning window based on respective feature descriptor of a pre-defmed dimension; and detect the vehicle based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers, which is selected based on the categorization of the vehicle into any of the commercial vehicle or the passenger vehicle and performs the detection based on ambient light conditions, wherein at least one classifier of the two or more classifiers is trained separately for detecting the vehicle in each region defined in the field of view of the host vehicle, and wherein in response to the detection, distance of the detected vehicle from the host vehicle is determined.

[00057] In an embodiment, the region in the field of view of the host vehicle is any of a far region, a middle region or a near region and wherein, the ROI is selected from High Definition (HD) resolution image for the far region, Video Graphics Array (VGA) resolution image for the middle region, and Quarter VGA (QVGA) resolution image for the near region.

[00058] In an embodiment, the processor further performs Non-maximal suppression (NMS) based on the categorization of the vehicle in order to provide an output of a single bounding box around the vehicle.

[00059] In an embodiment, the ambient light conditions are detected using state of head light switch of the host vehicle.

[00060] In an embodiment, the at least one classifier of the two or more classifier is trained using a training image dataset comprising training images, which are segregated based on resolution of each training image and the region defined in the field of view of the host vehicle. [00061] In an embodiment, training of the at least one classifier is performed using the training image dataset by: cropping one or more training images of each of the training image dataset; resizing each of the cropped one or more images to size of a nearest scanning window; removing undesirable structures from a classifier database of the at least one classifier; collecting false positives from the at least one classifier; and adding the false positives to the classifier database of a succeeding classifier.

[00062] In an embodiment, in response to detection of ambient light conditions pertaining to night, false positives from output of the cascade of two or more classifiers are removed using V channel extraction.

[00063] In an embodiment, position of the detected vehicle is determined based on width of the vehicle.

[00064] In an embodiment, the processor tracks the detected vehicle by using a tracker selected based on categorization of the vehicle, wherein the tracker determines a state by analysing detection of the vehicle in consecutive images of the set of images, the state being determined from any of an idle state, a pre-track state, a tracking state or a cancel state, and wherein the tracker provides history-based motion prediction to track motion of the vehicle in case detection of the vehicle is missed.

[00065] In an embodiment, the image sensor is a single Red Clear Clear Clear (RCCC) camera sensor.

[00066] Another aspect of the present disclosure provides a method for detecting a vehicle during day and night time, carried out according to instructions stored in a computer implemented in a host vehicle, comprising: receiving one or more images from an image sensor that images field of view of the host vehicle and defining a Region of Interest (ROI) for a set of images selected from the one or more images, wherein the ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the vehicle driver, and wherein a local histogram equalization-based contrast enhancement technique is performed in the ROI to make the output independent of illumination variation; determining a plurality of scanning windows in the ROI of each image based on categorization of the vehicle into any of a commercial vehicle or a passenger vehicle, wherein size of each scanning window is computed based on presumed height and width of the vehicle in the image; extracting Integral Histogram of oriented gradients (IHOG) features from each of the scanning window based on respective feature descriptor of a pre-defmed dimension; and detecting the vehicle, based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers, which is selected based on the categorization of the vehicle into any of the commercial vehicle or the passenger vehicle, wherein at least one classifier of the two or more classifiers is trained separately for detecting the vehicle in each region defined in the field of view of the host vehicle, and wherein in response to the detection, distance of the detected vehicle from the host vehicle is determined.

[00067] Those skilled in the art would appreciate that autonomous vehicle facilitate unmanned navigation, and braking in case of obstacles like pedestrian, car, heavy vehicles. If different camera sensors are used depending upon light conditions e.g. day or night, complications in the system are created. Therefore, the present disclosure provides an approach that uses single RCCC camera sensor to facilitate both day and night time vehicle detection. Further, detection of bottom-point and width-based distance correction of the vehicle is performed to have accurate distance of target vehicle (vehicle) from the host vehicle incorporating the system in order to facilitate smooth Automatic Emergency Braking (AEB) and Automatic Cruise Control (ACC). Also, cascaded Ada-boost and SVM classifier is used to provide optimized timing and efficient functionality. It would be appreciated that single SVM threshold used in conventional approach across the distance creates either false positives in near distance or miss detection issue in far region, to mitigate such issues; training for far, near and middle region is performed using different datasets. Also, a tracker with history based motion predictor is used to track vehicle motion in case of missing detections.

[00068] FIGs. 1A-B illustrate architecture of a vehicle detection system to illustrate its overall working in accordance with an embodiment of the present disclosure.

[00069] According to an embodiment, a vehicle detection system 100 (interchangeably referred to as system 100, hereinafter) is implemented in a host vehicle. The system 100 comprises an input unit 102, a processing unit 104 and an output unit 106. The input unit 102 may comprise an image sensor or a camera configured in the vehicle to capture images of field of view of the host vehicle. In an implementation, the image sensor or the camera may be placed below rear-view mirror of the host vehicle. The image sensor is a single mono-vision sensor or a RCCC camera sensor represented at block 156that facilitates detection during day and night. The processing unit 104 may comprise a processor and a memory and/or may be integrated with existing systems and controls of a host vehicle to form an advanced driver assistance system (ADAS), or augment an existing ADAS. For instance, signals generated by the processing unit 104 may be sent to engine control unit (ECU) of the host vehicle and may aid in parking of the host vehicle. The output unit 106 may be a display device or any other audio-visual device that provides waring to the driver when a vehicle is detected by the host vehicle.

[00070] According to an embodiment, during initialization and scanning window selection 108, the processing unit 104 receives one or more images from the input unit 102 or RCCC sensor 156and defines a ROI for a set of images selected from the one or more images. The ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the host vehicle. The region in the field of view of the host vehicle driver is any of a far region, a middle region or a near region such that the ROI is defined from HD resolution image for the far region, VGA resolution image for the middle region, and QVGA resolution image for the near region.

[00071] In an embodiment, RCCC sensor 156 use clear filters instead of blue and two green filters in 2x2 pixel neighbourhood along with a red filter. A clear filter has a concept similar to monochrome sensors. Thus, output generated by RCCC sensor 156 is almost as detailed as a monochrome output and still provides red colour information. At RCCC to YUV block 154, the output generated by the RCCC sensor 156 is converted to YUV format RCCC to YUV conversion may happen at camera side. Those skilled in the art would appreciate that RCCC camera sensor 156is preferred because it can be useful for both day and night time condition.

[00072] In an embodiment, camera parameter block 152, provides camera parameters including but not limited to, camera height (mm), camera pitch (degree), camera focal length (mm), sensor pixel size (micro m), etc. that are used for calibration by initialization block 158.The initialization block 158 initializes ROI parameter, scanning window parameter, classifier parameter, IHOG parameter, tracker parameter and stores computed results in constant arrays for further use. [00073] In an embodiment, at pre-processing block 160, a local histogram equalization- based contrast enhancement technique is performed in the ROI to make the output independent of illumination variation.

[00074] Furthermore, at scanning window generation block 166, the processing unit 104 determines a plurality of scanning windows in the ROI of each image based on categorization of the vehicle into any of a commercial vehicle or a passenger vehicle. Size of each scanning window is computed based on presumed height and width of the vehicle in the image

[00075] In an embodiment, during feature extraction 110, at IHOG feature extraction block 164, the processing unit 104 extracts IHOG features from each of the scanning window based on respective feature descriptor of a pre-defined dimension.

[00076] In an embodiment, during vehicle detection 112, the processing unit 104 detects the vehicle based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers, which is selected based on categorization of the vehicle into any of the commercial vehicle or the passenger vehicle e.g. an Ada-boost classifierl70 along with an SVM classifier 168 is used for commercial vehicle and an Ada-boost classifier 172 along with an SVM classifier 174 is used for passenger vehicle . The cascade of two or more classifiers performs the detection based on ambient light conditions, which are determined using state of head light switch of the vehicle. At least one classifier of the two or more classifiers is trained separately for detecting the vehicle in each region defined in the field of view of the host vehicle. Further, at least one classifier of the two or more classifier is trained using a training image dataset comprising training images, which are segregated based on resolution of each training image and the region defined in the field of view of the host vehicle. The training of the at least one classifier is performed using the training image dataset by: cropping one or more training images of each of the training image dataset; resizing each of the cropped one or more images to size of a nearest scanning window; removing undesirable structures from a classifier database of the at least one classifier; collecting false positives from the at least one classifier; and adding said false positives to the classifier database of a succeeding classifier.

[00077] In an embodiment, when the vehicle in detected, during false positive (FP) removal 114, at FP removal blocks 178 and 180, in response to detection of ambient light conditions (using day/night flag 196) pertaining to night, false positives from output of the cascade of two or more classifiers are removed using V channel extraction and feedback 176 for commercial and passenger vehicles respectively.

[00078] In an embodiment, during NMS 116, at NMS block 182 and 184 the processing unit 104 performs NMS to provide an output of a single bounding box around the detected vehicle for commercial and passenger vehicles respectively.

[00079] In an embodiment, during tracking 118, the processing unit 104 tracks the detected vehicle using tracker including tracker commercial 186 and tracker passenger 188based on categorization of the vehicle. The tracker determines a state by analysing detection of the vehicle in consecutive images of the set of images, the state being determined from any of an idle state, a pre-track state, a tracking state or a cancel state. The tracker also provides history-based motion prediction to track motion of the vehicle in case detection of the vehicle is missed. Further, at block 190, merging of commercial and passenger vehicle boxes in the image is performed.

[00080] In an embodiment, during distance computation 120, the processing unit 104 determines lateral and longitudinal distance of the detected vehicle from the host vehicle. At distance estimation block 194, position of the detected vehicle is determined based on width based distance correction 192.

[00081] FIG. 2 illustrates exemplary modules of a processing unit in accordance with an embodiment of the present disclosure.

[00082] In an aspect, the processing unit 104 may comprise one or more processor(s) 202. The one or more processor(s) 202 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and/or any devices that manipulate data based on operational instructions. Among other capabilities, the one or more processor(s) 202 are configured to fetch and execute computer-readable instructions stored in a memory 206 of the processing unit 104. The memory 206 may store one or more computer-readable instructions or routines, which may be fetched and executed to create or share the data units over a network service. The memory 206 may comprise any non-transitory storage device including, for example, volatile memory such as RAM, or non- volatile memory such as EPROM, flash memory, and the like. [00083] The processing unit 104 may also comprise an interface^) 204. The interface(s) 204 may comprise a variety of interfaces, for example, interfaces for data input and output devices, referred to as I/O devices, storage devices, and the like. The interface(s) 204 may facilitate communication of processing unit 104 with various devices coupled to the processing unit 104 such as the input unit 102 and the output unit 106. The interfaced) 204 may also provide a communication pathway for one or more components of the processing unit 104. Examples of such components include, but are not limited to, processing engine(s) 208 and data 210.

[00084] The processing engine(s) 208 may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine(s) 208. In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing engine(s) 208 may be processor executable instructions stored on a non-transitory machine-readable storage medium and the hardware for the processing engine(s) 208 may comprise a processing resource (for example, one or more processors), to execute such instructions. In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine(s) 208. In such examples, the processing unit 104 may comprise the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to processing unit 104 and the processing resource. In other examples, the processing engine(s) 208 may be implemented by electronic circuitry.

[00085] The data 210 may comprise data that is either stored or generated as a result of functionalities implemented by any of the components of the processing engine(s) 208.

[00086] In an exemplary embodiment, the processing engine(s) 208 may comprise an initialization and scanning window selection module 212, a feature extraction module 214, a vehicle detection module 216, a non-maximal suppression (NMS) module 218, a false positive removal module 220, a tracking module 222, a distance computation module 224 and other modules 226. [00087] It would be appreciated that modules being described are only exemplary modules and any other module or sub-module may be included as part of the system 100 or the processing unit 104. These modules too may be merged or divided into super-modules or sub-modules as may be configured.

Initialization and Scanning Window Selection Module 212

[00088] In an aspect, initialization and scanning window selection module 212 receives one or more images from the input unit and defines an ROI for a set of images selected from the one or more images. The ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the host vehicle. Further, the initialization and scanning window selection module 212 performs a local histogram equalization-based contrast enhancement technique in the ROI to make output independent of illumination variation. Furthermore, the initialization and scanning window selection module 212 determines a plurality of scanning windows in the ROI of each image based on categorization of the vehicle into any of a commercial vehicle or a passenger vehicle. The size of each scanning window is computed based on presumed height and width of the vehicle in the image.

[00089] In an embodiment, the region in the field of view of the host vehicle may be any of a far region, a middle region or a near region. The initialization and scanning window selection module 212 defines the ROI from HD resolution image for the far region, VGA resolution image for the middle region, and QVGA resolution image for the near region. FIG. 3A illustrates a block diagram of the initialization and scanning window selection module 212.

[00090] Those skilled in the art would appreciate that some computation, for example, setting parameters for various modules is one time, thus the initialization and scanning window selection module 212 performs such computation. In an embodiment, depending on detection range and range of vehicle height 302 along with camera configuration parameters 306 such as sensor pixel size pitch of the camera, initialization and scanning window selection module 212 initializes ROI parameter 304, scanning window parameter 308. Further, initialization and scanning window selection module 212 initializes other parameters such as classifier parameter 314, IHOG parameter 310, tracker parameter 312 and stores computed results in constant arrays for further use. [00091] In an embodiment, the plurality of scanning windows is determined in the ROI of each image based on categorization of the vehicle into any of the commercial vehicle or the passenger vehicle. Also, size of each scanning window is computed based on presumed height and width of the vehicle in the image. According to an example, passenger vehicles are detected from 5m to 90m whereas commercial vehicles are detected from 20m to 90m. As the distance of vehicle from the host vehicle increases, more details are required to extract features. In order to address such issues, scanning of the passenger vehicle is performed in three scales, i.e., HD, VGA and QVGA whereas scanning of commercial vehicle is performed in two scales, i.e., HD and VGA. Also, ROI for far distance vehicles is considered from HD (1280X960) image, ROI for detecting vehicles in the middle region of the image is taken from VGA (640X480) resolution image and ROI for near vehicles is taken from QVGA (320X240) resolution image.

[00092] FIGs. 3B, 3C and 3D illustrate scanning windows for the passenger vehicles in far, mid and near region respectively. As illustrated in FIGs. 3B-D, in order to detect passenger vehicles of different heights and widths, different window sizes are considered for scanning in three different scales. In an example, for detecting passenger vehicles in far region, 11 windows are taken in to consideration corresponding to HD image. Similarly, for detecting passenger vehicles in middle region, 1 1 windows are taken into consideration corresponding to VGA image and for detecting passenger vehicles in near region 10 windows are considered in QVGA region. Exemplary information of window size used for the three regions for detection of passenger vehicles is presented in Table 1.

[00093] Similarly, FIGs. 3E and 3F illustrate scanning windows of the commercial vehicles in far and middle region respectively. As illustrated in FIGs. 3E-F, in order to detect commercial vehicles of different heights and widths, different window sizes are considered for scanning in two different scales. In an example, for detecting commercial vehicles in far region, 11 windows are taken in to consideration corresponding to HD image. Similarly, for detecting commercial vehicles in middle region, 8 windows are taken into consideration corresponding to VGA image. Exemplary information of window size used for the two regions for detection of commercial vehicles is presented in Table 2.

Feature Extraction Module 214

[00094] In an aspect, feature extraction module 214 extracts IHOG features from each of the scanning window based on respective feature descriptor of a pro-defined dimension. Those skilled in the art would appreciate that a feature descriptor is a representation of a window that simplifies an image by extracting useful information and throwing away extraneous information. As an example, a window of size 56x112 of gray channel can have 56x112 values in terms of intensity values. These values are useful for viewing but it is of no use for the image recognition algorithms to efficiently classify the objects. A feature descriptor is used to generate an n- dimension feature vector to represent any window. The feature vector when fed into an image classification algorithms like SVM produce better results.

[00095] In an embodiment, an HOG is used as a feature descriptor where all scanning windows of different size are represented by corresponding feature vector of a dimension e.g.1548. Each window is divided into fixed number of cells, as size of the windows varies cell size of windows also varies. According to an example, in total 6600 windows can be obtained from 3 scales. Also, as can be noted, in FIGs. 3B-F many windows overlap with each other, in order to avoid repetitive HOG calculation for overlapping windows, an Integral HOG image is calculated individually for each scale that is used for calculation for windows. FIG. 4 illustrates a block diagram of the feature extraction module 214, where IHOG feature extraction 404-1, 404- 2. 404-N and cell size estimation 406-1, 406-2.. 406-N are performed on each scanning window of an input image 402 for window wise HOG feature calculation and normalization 408. Finally, 1548 HOG features are obtained for each scanning window at 410.

Vehicle Detection Module 216

[00096] In an exemplary embodiment, output of feature extraction module 214 is further used for classification and object detection by the vehicle detection module 216. In an aspect, vehicle detection module 216 detects the vehicle based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers that is selected based on categorization of the vehicle into any of the commercial vehicle or the passenger vehicle. The cascade of two or more classifiers performs the detection based on ambient light conditions.

[00097] According to an example, in total 5000 windows are there, classification stage decides in which of those windows a vehicle is present. Those skilled in the art would appreciate that SVM is a strong classifier that works on 1548 features, however, if only SVM is used then there would be a massive amount of computation. Therefore, in an example as illustrated in FIG. 5A an Ada-boost classifier 504 is used along with an SVM classifier 506. In context of the present example, Ada-boost classifier 504 includes Ada-boost based weak classifiers e.g. working on 100 features to reject large number of non-vehicle windows 510 in first stage.

[00098] In an example, IHOG feature 502 from each scanning window is fed in to the Ada-boost classifier 504. The Ada-boost classifier 504 has 100 weak classifiers so that the classifier has a capability to pass almost all the vehicles 508 and few non-vehicles 510. Windows classified as positive (with vehicles) from the Ada-boost classifier 504are fed into the SVM 506. As the images are passing through the Ada-boost classifier 504 prior to the SVM 506 the computational load on the SVM 506 is reduced. The SVM 506 performs final classification into vehicle windows 508 (positive) or non-vehicle windows 510 (negative). An exemplary output of the SVM classifier 506 is illustrated in FIG. 5B, which is a final decision from classification stage.

[00099] In an embodiment, the classifiers are trained using a training image dataset comprising training images, which are segregated based on resolution of each training image and the region defined in the field of view of the vehicle driver. In an example, vehicle and non- vehicle dataset is provided to train each classifier. Positive training dataset of the classifier is generated by iteratively training the classifier by inclusion of all the types of cars in equal numbers in case of passenger vehicle detection. Negative dataset is generated by taking the misclassified images of previous stage and other negatives like poles, road edges, building structures. Also, training dataset used for HD comprises those vehicles and non-vehicles which are obtained from HD scale and are at far distance. Similarly, training dataset used for VGA and QVGA regions comprises those vehicles and non-vehicles that are obtained from corresponding scales, middle and near region respectively. It would be appreciated that the technique disclosed herein differs from the convention approaches where same dataset is used across the scales that causes possibility of false positives in near region and also misses detection in far region.

[000100] In an example, as illustrated in FIG. 5C positive training dataset of the classifier is generated by iteratively training classifier by inclusion of all vehicles in equal numbers and negative dataset is generated by taking the misclassified images of previous stage of the classifier model. For example, positive dataset for Ada-boost classifier 552 is created by using images with vehicles with different back portion and negative dataset for Ada-boost classifier is created by considering images with non-vehicles such as trees, poles, road edges, pedestrians, buildings and structures, etc. Further, negatives obtained from detector 554 associated with the Ada-boost classifier 552, are added to the negative database of the SVM classifier 556. Additionally, false positives of the SVM classifier 556 are also added to the negative database of SVM classifier 556 for efficient training.

[000101] The training of the classifier is performed using training image dataset, which are segregated based on resolution of each training image and the region defined in the field of view of the host vehicle by cropping one or more training images, for example, passenger and commercial vehicles of various types are cropped and given to the classifier for training. Further, the cropped one or more images are resized to size of a nearest scanning window. Undesirable structures resembling vehicles are removed from the classifier database such that false positives from the classifier are collected and added to the negative classifier database of a succeeding classifier.

Non-Maximal Suppression (NMS) Module 218

[000102] Further, the vehicle detection module 216 is coupled with a Non-maximal suppression (NMS) module 218 that provides an output of a single bounding box around the detected vehicle. The output of the classifier as illustrated in FIG. 5B is multiple bounding boxes around the vehicle. The output of SVM classifier 506 is further fed to NMS module 218 to suppress the multiple bounding boxes and provide a single detection box around the vehicle based on confidence and location of centroid. Those skilled in the art would appreciate that, in an exemplary implementation, to reduce false positives to propagate in detection stage, only those vehicles are considered positive, which are surrounded by more than three detection windows. FTGs. 6A and 6B illustrate exemplary outputs of NMS module 218 for passenger and commercial vehicles respectively.

False-Positive Removal Module 220

[000103] In an embodiment, in response to detection of ambient light conditions pertaining to night, false positives from classifier output are removed using V channel extraction. For example, during night time some false positives are detected while extracting vehicles from the classifier using Y channel. In order to remove false detections V channel information is extracted. Therefore, during night time, as illustrated in FIG. 7B, output of NMS 702 is provided for V channel extraction 704 and thresholding 706 such that at block 708, the false positives are filtered out based on white pixel count of the image obtained after thresholding 706. FIG. 7A illustrates Y channel of the image, FIG. 7B illustrates V channel of the image and FIG.7C illustrates exemplary output after thresholding, from which the false detections can be filtered out. In an exemplary implementation, the false positive removal module 220 may receive classifier output from the vehicle detection module 216 such that NMS is performed subsequent to the false positive removal.

Tracking Module 222

[000104] In an aspect, tracking module 222 tracks the detected vehicle by using a tracker selected based on categorization of the vehicle. The tracker determines a state by analysing detection of the vehicle in consecutive images of the set of images such that the state is being determined from any of an idle state, a pre-track state, a tracking state or a cancel state. The tracker provides history-based motion prediction to track motion of the vehicle in case detection of said vehicle is missed.

[000105] The tracking module 222 uses a tracker for two purposes. Firstly, if some vehicle is continuously detected from classification stage and suddenly the detection is missed, the continuity of detection is maintained by the tracker. Secondly, the tracker keeps a control on propagation of false positives. An exemplary state diagram of the tracker is illustrated in FIG.8.

[000106] According to an example as illustrated in FIG. 8, initially, the tracker is in idle state. When the vehicle is detected first time, tracker enters in to pre- track state. In the Pre-track state tracker checks for detection of the same vehicle in the successive images. If detection is there then occurrence count is incremented, else missing count is incremented. The tracker enters into tracking state when the occurrence count exceeds a pre-track threshold. Conversely, if missing count exceeds a cancel track threshold in the pre-track state, the tracker enters to cancel track state. In the tracking state, the tracker checks for detection of the same vehicle in the successive images. If detection is there then the tracker continues in the same state. If matching object is not found, the tracker enters to cancel track state. In the cancel-track state also, the tracker checks for detection of the same vehicle in the successive images. If matching object is found, the tracker returns to tracking state, otherwise, missing count is incremented. When missing count exceeds a cancel track threshold, tracker enters back to idle state.

[000107] In an embodiment, when there is a detection miss, the tracker displays a flag indicating there is high probability of vehicle to be present at particular distance. Estimation of actual position of the vehicle is performed by distance computation module 224.

Distance Computation Module 224

[000108] Those skilled in the art would appreciate that it is necessary to know distance of the vehicle from the host vehicle in order to smoothly apply braking and avoid accidents. There are two components of the distance i.e. longitudinal component and lateral component.

[000109] Longitudinal distance is the distance of the vehicle in longitudinal direction from the host vehicle. Due to some fluctuation in detections, longitudinal distance varies in each consecutive image/frame. To avoid the fluctuation in longitudinal distance, width based correction method and low pass filtering are performed to correct the distance that is close to match real world distance.

[000110] As illustrated in FIG. 9, at block 902, width of the vehicle in the image is found and at block 904, it is determined whether the width is in certain limits. If not, at block 906 raw distance is considered. If width is in certain limits, at block 908 correct distance is determined using look up table 910. As, raw distance does not match with real world distance most of times, distance correction is performed based on width. Thus, width of the vehicle is calculated and if the width of the vehicle is within limits, distance is determined using look up table distance. Lateral Distance is distance of the vehicle in lateral direction from the host vehicle centre. Due to some fluctuation in detections, lateral distance varies in each frame. To avoid the fluctuations in lateral distance, a low pass filter has been used.

Other Modules 226

[000111] In an aspect, other modules 226 implement functionalities that supplement applications or functions performed by the system 100, processing unit 104 or the processing engine(s) 208.

[000112] Although the proposed system has been elaborated as above to include all the main modules, it is completely possible that actual implementations may include only a part of the proposed modules or a combination of those or a division of those into sub-modules in various combinations across multiple devices that may be operatively coupled with each other, including in the cloud. Further the modules may be configured in any sequence to achieve objectives elaborated. Also, it may be appreciated that proposed system may be configured in a computing device or across a plurality of computing devices operatively connected with each other, wherein the computing devices may be any of a computer, a smart device, an Interet enabled mobile device and the like. Therefore, all possible modifications, implementations and embodiments of where and how the proposed system is configured are well within the scope of the present invention.

[000113] FIG. 10 is a flow diagram illustrating a method for detection of a vehicle in accordance with an embodiment of the present disclosure.

[000114] In an aspect, the proposed method may be described in general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, functions, etc., that perform particular functions or implement particular abstract data types. The method can also be practiced in a distributed computing environment where functions are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, computer executable instructions may be located in both local and remote computer storage media, including memory storage devices.

[000115] The order in which the method as described is not intended to be construed as a limitation, and any number of the described method blocks may be combined in any order to implement the method or alternate methods. Additionally, individual blocks may be deleted from the method without departing from the spirit and scope of the subject matter described herein. Furthermore, the method may be implemented in any suitable hardware, software, firmware, or combination thereof. However, for ease of explanation, in the embodiments described below, the method may be considered to be implemented in the above described system.

[000116] In an aspect, present disclosure elaborates upon a method for vehicle detection that comprises, at step 1002, receiving one or more images from an image sensor that images field of view of the host vehicle and defining a Region of Interest (ROI) for a set of images selected from the one or more images, wherein the ROI is defined in each image of the set of images based on resolution of each image and a region defined in the field of view of the vehicle driver, and wherein a local histogram equalization-based contrast enhancement technique is performed in the ROI to make the output independent of illumination variation, and at step 1004, determining a plurality of scanning windows in the ROI of each image based on categorization of the vehicle into any of a commercial vehicle or a passenger vehicle, wherein size of each scanning window is computed based on presumed height and width of the vehicle in the image.

[000117J The method further comprises, at step 1006, extracting Integral Histogram of oriented gradients (IHOG) features from each of the scanning window based on respective feature descriptor of a pre-defined dimension and at step 1008, detecting the vehicle, based on the extracted IHOG features from each scanning window using a cascade of two or more classifiers, which is selected based on said categorization of the vehicle into any of the commercial vehicle or the passenger vehicle, wherein at least one classifier of the two or more classifiers is trained separately for detecting the vehicle in each region defined in the field of view of the host vehicle, and wherein in response to said detection, distance of the detected vehicle from the host vehicle is determined.

[000118] As would be readily appreciated, while primary application for disclosure as elaborated herein is in the automotive domain for vehicle detection, it may be used in non- automotive domain as well wherein any moving object may be similarly detected.

[000119] As elaborated above, the proposed system incorporates various techniques, which provide various advantages over existing methodologies. For example, the proposed system uses a single RCCC camera sensor for detecting the vehicle in day and night time. Further, the vehicles are characterized into commercial and passenger vehicle in day and night The day or night is determined from host vehicle beam such that the classifier is tuned for day/night Further, extraction of V channel information is performed for removal of false positive in night time and width based distance estimation is performed by correcting the distance based on width of the vehicle. The proposed system also estimates vehicle trajectory by computing lateral and longitudinal distance of the vehicle from the host vehicle. Furthermore, tracker with history based motion predictor is used to track vehicle motion in case of missing detections. Exemplary Test Results

[000120] Those skilled in the art would appreciate that conventional approaches use IR camera sensor for night time and RGB camera sensor for day time. Embodiments of the present disclosure use a single camera sensor and hence, reduce system complexity. Further, as illustrated in FTGs. 11 A-C provide comparison of results obtained using system of the present disclosure (right side) with results obtained using a conventional approach (left side). Bounding box obtained using the conventional approach was highly unstable which intern gave inaccurate distance. The present system provided a stabilized bounding box and accurate vehicle distance that aids to deal with smoother operation in complex conditions while vehicle navigating in complete autonomous or semi- autonomous mode of ACC or AEB control.

[000121] FIGs. 11 A-C provide testing results when same classifier is used across the regions on the left side and when different Ada-boost and SVM Classifier in far, middle and near region in accordance with an embodiment of the present disclosure on the right side respectively. As would be appreciated detection in all regions and specifically in far region has been significantly improved after using embodiments herein over using same classifier across the distance.

[000122] As used herein, and unless the context dictates otherwise, the term“coupled to” is intended to include both direct coupling (in which two elements that are coupled to each other or in contact each other)and indirect coupling (in which at least one additional element is located between the two elements). Therefore, the terms“coupled to" and“coupled with” are used synonymously. Within the context of this document terms“coupled to” and“coupled with" are also used euphemistically to mean“communicatively coupled with” over a network, where two or more devices are able to exchange data with each other over the network, possibly via one or more intermediary device.

[000123] Moreover, in interpreting both the specification and the claims, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms “comprises” and“comprising” should be interpreted as referring to elements, components, or steps in a non-exclusive manner, indicating that the referenced elements, components, or steps may be present, or utilized, or combined with other elements, components, or steps that are not expressly referenced. Where the specification claims refers to at least one of something selected from the group consisting of A, B, C ....and N, the text should be interpreted as requiring only one element from the group, not A plus N, or B plus N, etc.

[000124] While some embodiments of the present disclosure have been illustrated and described, those are completely exemplary in nature. The disclosure is not limited to the embodiments as elaborated herein only and it would be apparent to those skilled in the art that numerous modifications besides those already described are possible without departing from the inventive concepts herein. All such modifications, changes, variations, substitutions, and equivalents are completely within the scope of the present disclosure. The inventive subject matter, therefore, is not to be restricted except in the spirit of the appended claims.

ADVANTAGES OF THE PRESENT DISCLOSURE

[000125] The present disclosure provides a system and method for detecting a vehicle during day and night time.

[000126] The present disclosure provides a system and method for detecting a vehicle that provides appropriate warnings on detection.

[000127] The present disclosure provides a system and method for detecting a vehicle that tracks unpredictable motion of the vehicle.

[000128] The present disclosure provides a system and method for detecting distance of a vehicle for smoother Automatic Emergency Braking (AEB) and Automatic Cruise Control (ACC) in complex scenarios.

[000129] The present disclosure provides a system and method for detecting vehicles with different back portions.

[000130] The present disclosure provides a robust, economic and simple system and method that accurately detects a vehicle.