FIELD

Cross-Reference To Related Application [0001] This application claims the benefit of U.S. Provisional Application Serial No. 63/022,935, filed May 11, 2020, the disclosure of which is hereby incorporated by reference in its entirety.

Field of Invention

[0002] The present disclosure relates generally to identifying the male components of selected plants in a field. More particularly, the present disclosure relates to determining and locating if there are unwanted tassels (i.e., male components) of com plants in a hybrid seed com field. More particularly still, the present disclosure determines and locates the existence of unwanted tassels in a hybrid seed com field after a detasseling operation by imaging the com plants and identifying the tassels in the images.

Background

[0003] Hybrid com may be developed through cross-pollination of the silks (female) of one variety of com with the tassels (male) of another. Such cross-pollination produces hybrid com that often provides superior characteristics such as higher yield, improved resistance to drought and/or disease, and other advantageous characteristics. In general, in order to cross-pollinate different varieties, the tassels of first selected com plants are removed while leaving the female silk intact. The tassels of second selected com plants (i.e., from a different variety) are used to cross-pollinate the female parts of the first selected plants.

[0004] To remove the tassels from the first selected plants, the tassel is cut and/or pulled from the top of the com plant. Traditionally, this process is performed by passing a machine, potentially multiple times, through the first selected com plants. Crews then make additional passes to clean up any left behind unwanted tassels from the first selected com plants. However, tassels are left on the second selected com plants to provide the necessary cross-pollination. In order to facilitate the process of removing the unwanted tassels from first selected plants and not from second selected plants, fields are planted in a particular pattern. The pattern establishes known locations/rows of the first and second selected plants. For example, a field may be planted in a pattern with one row of male plants (second selected plants) and then four rows of female plants (first selected plants), followed by another row of male plants and then four rows of female plants, and so on in a repeating pattern.

[0005] Seeds from the first selected com plants that have been cross-pollinated are ultimately harvested as a hybrid seed and used for production com. To achieve the desired purity for a hybrid seed crop, a high percentage of the tassels from the first selected plants must be removed. Further, to maximize seed com yield, machines removing unwanted tassels must minimize the amount of the com plant that is removed. It will also be appreciated that the time window in which such tassels must be removed is very short and labor intensive — and therefore expensive.

[0006] By way of example, in many cases, the desired purity is 99.5% - 99.9% for seed. This level of purity requires that fewer than 5 unwanted tassels are found in a 1000 of first selected com plants. The reason for this level of purity is due to one unwanted tassel may create -200 unwanted seeds in production com. If the unwanted seeds make it into bags of seed com, the unwanted seeds will not have the desired characteristics of the desired seed com.

[0007] Mechanical devices used to remove tassels from the first selected com plants (“detasselers”) have been developed. As noted above, to obtain the desired purity of the hybrid seed, a high removal percentage of tassels must be achieved from the first selected com plants. But there are several issues that must be overcome in order to achieve the necessary percentages, including varying heights of the com plants, tassels missed by the detassel machine, tassels which are removed but do not fall to the ground (referred to as “hangers” herein), and rogue or unwanted com plants, among others, all of which are referred to herein as “unwanted” tassels.

[0008] Therefore, there is a need in the art to determine when and where these issues arise. The present invention addresses this need, as well as other problems associated with automated com detasseling. Summary

[0009] A preferred embodiment of a method, apparatus and system constructed according to the principles of the present invention includes a system which may be adapted for use directly on a detasseling machine moving through a field of com. Alternatively, it may be used on a machine moving through the field subsequent to a detasseling operation having occurred. Still further, data may be collected via either of the two previous alternatives with the processing occurring remotely and/or later in time. In any of these alternatives, unwanted tassels from first identified plants - e.g., which have been missed or are hangers — are identified and mapped. By doing so, the unwanted tassels can be properly dealt with in real time or with another pass through the field. By eliminating such unwanted tassels, the desired cross-pollinating percentages are increased.

[0010] Imaging devices are used to view and/or record images of the at least first selected plants. A GPS or other geo-location identifier is associated with each image. The images are then provided to a computing system to determine if the images include unwanted tassels. The computer system may use a machine learning program to identify any such unwanted tassels. In this manner when such a tassel is identified, the associated GPS coordinates provide for real-time, near real-time, or later automated re-entry into the field with equipment to remove the identified unwanted tassel. Alternatively, the mapped unwanted tassels might be located and removed manually, with the mapping function providing a solution that accomplishes the removal in less time and more completely than previous methods.

[0011] Therefore, according to one aspect of the invention, there is provided a method for detecting and identifying objects of interest relative to first identified plants growing in a field, of the type used for cross-pollination of seed crops, the method comprising: collecting images of the plants after removing a sexual part of the first identified plants; applying a trained model to identify objects of interest in the images; associating a physical location marker to each identified object of interest; and outputting a file of the combined objects of interest and corresponding physical locations.

[0012] There is also provided in accordance with the preceding paragraph, a method where the sexual part is a male part; where the first identified plant is com, the sexual part is a tassel, and the objects of interest are unwanted tassels; and/or where unwanted tassels include tassels that were not removed by being missed, hanger tassels, not removed by the plant being too short, and tassels from rogue plants. Further, in accordance with the preceding paragraph, the collected images may be photographs taken by cameras mounted on equipment moving through the field; where there are a plurality of forward facing cameras pointed in the forward direction of travel of the equipment (and/or one or more of the plurality being generally forward facing) and a plurality of rearward facing cameras pointed in the rearward direction of travel of the equipment (and/or one of the plurality being generally rearward facing), whereby the likelihood of a tassel being missed due to being covered by plants or leaves is minimized; and/or wherein there are two forward facing cameras and three rearward facing cameras.

[0013] According to another aspect of the invention, there is provided a method for training and applying a model to detect and identify anomalies located in plants growing in a field, of the type used for cross-pollination of seed crops, the method comprising: training the model using images with identified male parts of the plants; removing the male parts of a first identified group of plants; collecting images of the plants after the removal step; applying the trained model to identify anomalies in the first identified group of plants; associating a physical location marker to each identified anomaly; and outputting a file of the combined anomaly and physical location.

[0014] According to a still further aspect of the invention, there is provided a system applying a model to detect and identify objects of interest located in plants growing in a field, of the type used for cross-pollination of seed crops, comprising: a memory structure, the memory structure including information for a machine trained model to identify objects of interest; a plurality of cameras for imaging the plants, the plurality of cameras outputting a corresponding plurality of image signals including images of the plants; a GPS device arranged and configured to provide a geo-locational signal; and a processor operatively connected to the plurality of cameras, the GPS device and the memory structure, the processor arranged and configured to review the plurality of image signals with a machine trained model and the information for the machine trained model to identify the objects of interest included in the images and to correlate the object of interest with a geo-locational position. [0015] While the invention will be described with respect to preferred embodiment configurations and with respect to particular components used therein, it will be understood that the invention is not to be construed as limited in any manner by either such configuration, type of imaging system used, machine learning program, or geo location systems. Instead, the principles of this invention extend to an application which automatically determines the existence of male or female parts of plants, and more specifically unwanted tassels of com plants, and associates the same with a location identifier.

[0016] Further, while use of the system in connection with cross-pollinating com is described herein, the type of plant may vary. For example, the system may be employed with other seed crops such as wheat, soybeans, and sorghum. Still further, while programming steps and logical flow will be described herein in connection with a computer system used with an embodiment, it will be understood that the scope of the invention is not to be so limited. The invention may be employed in other environments in which the information is reviewed. These and other variations of the invention will become apparent to those skilled in the art upon a more detailed description of the invention.

[0017] The advantages and features which characterize the invention are pointed out with particularity in the claims annexed hereto and forming a part hereof. For a better understanding of the invention, however, reference should be had to the drawings which form a part hereof and to the accompanying descriptive matter, in which there is illustrated and described a preferred embodiment of the invention.

Brief Description of Drawings

[0018] Referring to the drawings, wherein like numerals represent like parts throughout the several views:

[0019] FIG. 1 is a functional block diagram illustrating the various components of an embodiment used in connection with identifying anomalies or objects of interest in a field row that includes first identified plants in accordance with the principles of the present invention. [0020] FIG. 2 is an illustration of an adjustable tool bar mounted on equipment with a prime mover used in connection with seed com fields and including a plurality of detasseler devices cooperatively connected to the tool bar.

[0021] FIG. 3a is a schematic diagram illustrating the location of cameras in an embodiment mounted on the tool bar 106 of FIG. 2.

[0022] FIG. 3b is a schematic diagram illustrating in some embodiments an optional vertical adjustment of the cameras of FIG. 3a.

[0023] FIG. 3c is a schematic of a portion of a tool bar 106 illustrating an example of the locations of the forward and rearward facing cameras.

[0024] FIG. 4 illustrates an example block diagram of a computing system that may be employed to implement aspects of the system of FIGs. 1 and 7.

[0025] FIG. 5 illustrates a method for generating and training a model to detect and map unwanted tassels, hangers, shorts and rogue plants (i.e., objects of interest or anomalies).

[0026] FIG. 6a illustrates a representative spreadsheet output of the tassel anomalies identified by use of the present invention.

[0027] FIG. 6b illustrates a representative map output of the tassel anomalies identified by use of the present invention.

[0028] FIG. 7 illustrates functional blocks of a system which may be used to practice the operation of the system in accordance with principles of the invention.

Detailed Description

[0029] The principles of the present invention apply particularly well to its application in a method, apparatus and system useful in reducing the amount of manual labor associated with current processes to locate and eliminate unwanted com tassels, and which may aid in increasing the percentage of cross-pollinated seed com. However, it will be appreciated by those of skill in the art that the type of plant may vary. For example, the system may be employed with other seed crops such as wheat, soybeans, and sorghum. Further, while the type of image collection device, programming steps and logical flow are presented herein in connection with an embodiment, it will be understood that the scope of the invention is not to be so limited. The invention may be employed in other cross-pollination environments in which the presence of male or female components of the plants are determined and mapped. [0030] Turning now to Fig. 1, there is illustrated an embodiment vision and processing system generally at 10. The system 10 includes a vision and processing computer 250 (described in more detail further below). The computer 250 receives input from a plurality of rear facing cameras 29a, 29b and 29c; a plurality of forward facing cameras 30a and 30b; GPS coordinates from a GPS receiver 26; and a display and input device(s) 27. Optionally, a telematics device 28 (or other over the air transmission of data including cellular, radio, Bluetooth, and other near point transmission protocols and devices) may be included for remote monitoring. The computer 250 includes a machine learning program to receive images from the cameras 29 and 30 to identify unwanted tassels on first selected plants (i.e., that should have been cut or pulled, did not fall to the ground and/or are located on rogue plants).

[0031] In the embodiment illustrated in herein, cameras using visually perceptible wavelengths are illustrated and described. However, it will be appreciated by those of skill in the art that image collection devices utilizing other wavelengths, including non visual wavelengths, may be utilized. In addition, while the cameras are described as forward and rearward facing, such cameras may also be mounted at angles relative to those directions. Also, cameras may optionally be employed having a view that is normal (or an angle relatively normal) to a forward direction.

[0032] Further, aspects of the invention include training a model to recognize features within images captured by the cameras in a subject field. In an embodiment, an output file is created from the recognized features of those tassels which are:

1. missed first identified plant tassels (e.g., tassels which should have been cut or pulled, but which were missed in the pass through the field with the detasseler equipment);

2. hangers (e.g., pull tassels which are “hanging” on the leaves above the silk (female part) of the first selected plants);

3. shorts (i.e., those tassels which were missed being cut because of the shorter height of the com plant); and

4. Rogue plants (i.e., unwanted plants).

[0033] With regard to rogue plants, in the context of the present example embodiment, they may be defined to be com plants which are: A. the wrong seed which was planted in the rows that are made up of the first selected plants. These may be determined as they generally exhibit a different color green and are typically much taller than the surrounding plants;

B. volunteer com which are seeds that are from a previous year’s crop or in some situations from events such as a bird dropping a seed. Many times these particular plants are not in either of the currently planted rows.

C. Rogue plants A and B are typically the types of rogue plants found by existing processes (e.g., such as manually removed by a crew moving through the field). However, to find all of the rogues, the leaf structure of each plant would need to be examined. Embodiments constructed in accordance with the principles of the present invention could be trained to perform this task, which would locate additional rogue plants, and which would lead to increased purity.

[0034] Also included in the output file is an associated GPS position (or other geo- locational data) of the unwanted tassel and, optionally, a vertical elevation. The output file may be in a spreadsheet format (e.g., such as EXCEL)(best seen in FIG. 5a) and/or presented as a map of the field (e.g., in XML format)(best seen in FIG. 5b).

[0035] A flowchart for illustrating a machine learning process is described further below in connection with FIG. 6. However, a brief overview of the process is next provided for background and context. In an embodiment, a model for recognizing the male and female parts of the com plant is initially trained by collecting images of tassels and/or silks, viewing the images, and labeling the tassels and silks appropriately. The labeled images are then provided to a machine learning algorithm to train the algorithm to identify tassels and silks in subsequent captured images (i.e., by cameras moving through a filed after a detasseling operation).

[0036] Accordingly, the disclosed approach identifies tassels from first identified com plants that should have been cut or pulled but missed, were cut or pulled but did not fall to the ground, should have been cut or pulled but were on shorter plants and are on rogue plants. Each of these specific failures to remove tassels are created as a separate record in the output file identified above. However, as used herein they are collectively referred to as anomalies or objects of importance (i.e., an item, event, or observation that does not conform to an expected result). By identifying and outputting only the anomalies or objects of importance, the present system outputs only data in what would otherwise be a large dataset. Further, the output record leads to saved costs by pinpointing where the anomalies are located in the fields.

[0037] Turning now to Fig. 2 a detassel er system is illustrated generally at 100. The system includes a chassis 102 supporting a detasseling head 114. The head 114 includes a toolbar 106 supporting row units 200 each of which include pullers or cutters. The toolbar 106 is mounted on a supporting linkage to adjust the height of the toolbar supporting the row units 200. Depending on the number of row units 200 supported and the width of the toolbar 106, the toolbar 106 may have multiple sections including a center section and folding outer sections (wings) for travel and storage. The chassis 102 is typically supported on four wheels 110 and includes a cab and a motor. It can be appreciated that the chassis 102 may be configured for use in applications other than detasseling such as supporting sprayers or other agricultural implements. The detasseling head 114 is therefore interchangeable with other types of heads performing other agricultural tasks. It can be appreciated that the wheels 110 help elevate and maintain the chassis 102 high off the ground so that it may drive over tall plants such as com and cause minimal damage to the field. The general forward direction of travel of the system 100 during operation is illustrated by the designation “X” shown in FIG. 3 a.

[0038] The detasseling head 114 includes both the toolbar support linkage and the toolbar 106. The toolbar support linkage provides for adjusting the height of the entire toolbar 106. In some embodiments, row units 200 include two cutters and are individually adjustable for more precise and more efficient detasseling. Hydraulic cylinders extend and retract to raise and lower a linkage cooperatively connecting the chassis 102 to the detasseling head 114. Therefore, by extending and retracting the hydraulic cylinders, the upper links and lower links pivot upward and downward and therefore adjust the overall height of the toolbar 106.

[0039] In addition, each pair of row units 200, independently move up and down hydraulically and can be controlled by either photocell or laser height control system mounted at the front of the row unit 200. In the case of a photocell system, a pair of vertically spaced photocells are mounted at a level corresponding to the height of the tops of the com. If both photocells are blocked, then the unit is too low. Alternatively, if neither photocell is blocked then the unit is too high. Accordingly, the system adjusts the row unit 200 whereby the top photocell is not blocked and the bottom photocell is blocked. In the case of a laser height adjustment system, a laser is used to determine the distance above the ground in order to maintain an appropriate height of the row unit over undulating ground. In either case, the systems operate to maintain a more optimum vertical height of the row unit 200 relative to the varying heights of the first selected tassels.

[0040] Turning to FIGs. 3 a, 3b and 3 c, the location of rear facing cameras 29a, 29b, and 29c (collectively cameras 29) and forward facing cameras 30a and 30b (collectively cameras 30) are shown schematically mounted on toolbar 106. Mounting bar 109 may be connected to toolbar 106 to provide a range of mounting heights for the cameras 29 and 30 between a lowered position designated 29a’ and an upper position designated 29a (best seen in Fig. 3b). The toolbar 106 is connected to chassis 102 via a linkage. Mounted on the rear of the chassis 102 is an optional toolbar 107. The optional toolbar 107 provides a location for the optional cameras 31a and 31b. The optional toolbar 107 may also include a mounting bar to provide vertical mounting options for the cameras 31. Cameras may also be mounted above and/or under the top canopy of the crop between the rows looking either forward, rearward, angles relative to forward and rearward and/or sideways.

[0041] Thus, cameras 29 and cameras 30 are used to look rearward and forward, respectively, relative to the direction X shown in Fig. 3a. It is believed that having images of the selected first plants (and adjoining rows or more) from both directions is important in locating the unwanted tassels. For example, including unwanted tassels in some images is made difficult because the surrounding plants “cover up” some of the plants — especially the “shorts” or smaller plants. Accordingly, as shown in FIG. 3c, in some embodiments there are two forward facing cameras 30a and 30b and three rear facing cameras 29a, 29b, and 29c for each row including first selected com plants of interest.

[0042] It will be appreciated that the number and type of cameras, lens and light filters may vary depending on the specific application and detasseling equipment. Further, the advantageous height for the cameras 29 and 30 may vary depending on the height of the crop, number of leaves, and size of leaves, among other factors. Still further, lighting may be provided to assist in the collection of the images. [0043] According to some embodiments, a variety of cameras and lenses may be used to detect unwanted tassels. For example, using a specialized camera or lens may assist in detecting unwanted tassels - which may reduce the required computing power needed to detect unwanted tassels. More specifically, it may be beneficial to utilize cameras that detect different light spectrums (e.g., thermal or temperature measuring cameras). Also, lenses which filter the light spectrum may be used to assist in a Normalize Difference Vegetation Index (NDVI) or similar calculation. A camera/lens system providing a spectrum for NDVI typically provides red and Near Infrared (NIR) in accordance with the following equation (1):

(1 ) NDVI = (NIR - Red) / (NIR + Red) where each value is the light energy for given spectrum of a given pixel. It is believed that this system would enable the chlorophyll in a plant to be highlighted. Since a tassel has much less chlorophyll than a leaf, an unwanted tassel may be highlighted in such a system against the plants, stalks and leaves containing more chlorophyll.

[0044] Now turning to FIG. 4, an example block diagram of a computing system 250 that may be used to implement aspects of the system 10 of FIG. 1 is illustrated. The computing system 250 can be used to implement, for example, the identification of the anomalies. In further aspects, the computing system 250 can represent the computing device used to train the machine learning program.

[0045] In the embodiment shown, the computing system 250 includes at least one central processing unit (“CPU”) 201 and graphics processing units (“GPU”) 202, a system memory 208, and a system bus 222 that couples the system memory 208 to the CPU 201 and GPU 202. The system memory 208 includes a random access memory (“RAM”) 210 and a read-only memory (“ROM”) 212. A basic input/output system that contains the basic routines that help to transfer information between elements within the computing system 250, such as during startup, is stored in the ROM 212. The computing system 250 further includes a mass storage device 214. The mass storage device 214 is able to store software instructions and data.

[0046] It will be appreciated that due to the large number of images being processed to locate the anomalies/objects of interest in embodiments of the present invention, a GPU 202 is useful in order to rapidly manipulate and process the images. The GPU’s highly parallel structure makes it more efficient than general-purpose central processing units (CPUs)(i.e., for algorithms that process large blocks of data in parallel). Thus, inclusion of a GPU 202 aids in the efficiency and realization of embodiments.

[0047] The mass storage device 214 is connected to the CPU 201 and GPU 202 through a mass storage controller (not shown) connected to the system bus 222. The mass storage device 214 and its associated computer-readable data storage media provide non-volatile, non-transitory storage for the computing system 250. Although the description of computer-readable data storage media contained herein refers to a mass storage device, such as a hard disk or solid-state disk, it should be appreciated by those skilled in the art that computer-readable data storage media can be any available non-transitory, physical device or article of manufacture from which the central display station can read data and/or instructions.

[0048] Computer-readable data storage media include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable software instructions, data structures, program modules or other data. Example types of computer-readable data storage media include, but are not limited to, RAM, ROM, EPROM, EEPROM, flash memory or other solid state memory technology, CD-ROMs, digital versatile discs (“DVDs”), other optical storage media, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by the computing system 250.

[0049] According to various embodiments of the invention, the computing system 250 may operate in a networked environment using logical connections to remote network devices through a network 260, such as a wireless network, the Intranet, the Internet, or another type of network. The computing system 250 may connect to the network 260 through a network interface unit 204 connected to the system bus 222. It should be appreciated that the network interface unit 204 may also be utilized to connect to other types of networks and remote computing systems. The computing system 250 also includes an input/output controller 206 for receiving and processing input/output from a number of other devices, including a touch user interface display screen, keyboard and/or other type of input/output device (collectively designated 27).

[0050] As mentioned briefly above, the mass storage device 214 and the RAM 210 of the computing system 250 can store software instructions and data. The software instructions include an operating system 218 suitable for controlling the operation of the computing system 250. The mass storage device 214 and/or the RAM 210 also store software instructions, that when executed by the CPU 201 and GPU 202, cause the computing system 250 to provide the functionality of identifying anomalies/objects of interest, pairing the individual anomaly with a GPS coordinate and creating a record of the same.

[0051] Turning now to FIG. 5 a method for generating and training a machine model to detect and map unwanted tassels and hangers (“anomalies” or “objects of interest”) is shown generally at 500.

[0052] First at 501, images are collected for the purpose of training the system to recognize tassels. In an embodiment, an imaging processing apparatus may be used for the purpose of collecting images. The images are then utilized to determine machine performance as it relates to a detasseling operation.

[0053] At block 502, the system utilizes machine learning methods to create an inference model that detects the presence of objects to be classified and classifies the objects of importance in the image. The objects of importance are a tassel that has not been detasseled, a tassel that has been detasseled but is still on a com plant and has not reached the ground, or a rogue plant. In a preferred embodiment, the software is constructed to conduct the tassel and rogue plant detection in real-time so that the operator of the detasseling equipment 102 has up to date information on performance vis-a-vis the detasseling operation. In another embodiment, the images are captured, saved, and evaluated at a later time to determine the operation performance in relation to the detasseling operation. The objects of importance are logged in an operation performance record.

[0054] In a preferred embodiment the system utilizes a neural network to classify the objects of importance in the image. However, it is understood that other techniques such as Naive Bayes, Linear SVM, Logistic Regression, or other machine learning techniques could be utilized for the model. Further, the system identifies the geographic location of the objects of importance and records that information as part of the operation performance record.

[0055] At block 503, the system tracks objects of importance between concurrent images to eliminate double counting of objects of importance. Vehicle speed information from a vehicle speed sensor and/or data from a GPS unit could be utilized by the system to aid in determining if an object of importance has been previously identified. One manner of determining this is by utilizing the speed information and knowledge of the camera field of view to calculate where the object of importance should be in consecutive collected images. If subsequent images contain an object of importance and that object is in a location that matches where a previous object of importance was projected to be, the system will determine that it is the same object and not count that object again. A person trained in the art will recognize that there are alternative methods to avoid double counting objects of importance.

[0056] At block 504 the system compiles information of the identified objects of interest. By doing so, a summary may be created of the machine performance for the detasseling operation. In addition, a table or map of the objects of interest (in this example unwanted tassels) may be created. Examples of such a table or map are illustrated in FIGs. 6a and 6b.

[0057] Turning now to FIG. 7 a schematic diagram of an embodiment for use in the field is shown generally at 10. An imaging processing computer 250 is included as described above. A GPS device 26 provides location data to the computer 250 and a vehicle speed sensor 25 provides the velocity over the ground to the computer 250. The velocity over the ground may assist the system in determining duplicates in images as noted above. A user or human interface 27 is connected to the computer 250 to provide input, output, and system status. Input from cameras 29 and 30 are provided to the computer 250. As shown in FIG. 7, there are a plurality of cameras 29a through 29n and 30a through 30n.

[0058] The final number of cameras is dependent on the desired number and angles of images to be taken of each row including first selected and/or other rows in which objects of interest are being found. By way of example, for the equipment 100 shown in FIG. 2 and FIG 3a, there are six detassel units 200. If five cameras (e.g., two forward and 3 rearward) are used in connection with each set of detassel units 200, then 15 cameras in total might be included in the system 10.

[0059] Also shown in FIG. 7 is a remote or desktop computer 250’ which may be connected to the image processing computer 250 to review real time data of the system 10 and/or to provide analysis and processing of generated maps.

[0060] It will be appreciated that the principles of this invention apply broadly to a method, apparatus and system of identifying unwanted tassels in a field of seed com.

While particular embodiments of the invention have been described with respect to its application, it will be understood by those skilled in the art that the invention is not limited by such application or embodiment or the particular components disclosed and described herein. It will be appreciated by those skilled in the art that other components and modules that embody the principles of this invention and other applications therefor other than as described herein can be configured within the spirit and intent of this invention. The arrangement described herein is provided as only one example of an embodiment that incorporates and practices the principles of this invention. Other modifications and alterations are well within the knowledge of those skilled in the art and are to be included within the broad scope of the appended claims.