Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
AN ARTICLE INDICATION SYSTEM
Document Type and Number:
WIPO Patent Application WO/2022/005305
Kind Code:
A1
Abstract:
A system for indicating articles to a manual grader during a manual grading operation comprises an upstream inspection device, and processor and a downstream indication device. The upstream inspection device comprises an image capture device to capture an inspection image of the plurality of articles. The processor is configured to detect one or more characteristics of the articles based on the inspection image. The downstream indication device comprises a projection device configured to project a visual indication onto each article on the conveyor comprising the one or more characteristics to highlight each article comprising the one or more characteristics to a manual grader during manual grading of the plurality of articles.

Inventors:
KNIGHTLEY SIMON (NZ)
Application Number:
PCT/NZ2021/050099
Publication Date:
January 06, 2022
Filing Date:
June 25, 2021
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
COMPAC TECH LIMITED (NZ)
International Classes:
B07C7/00; B07C5/342; B07C7/04; G01N21/89; G06T7/00; H04N5/74
Domestic Patent References:
WO2019151394A12019-08-08
Foreign References:
CA2863566A12016-03-12
Attorney, Agent or Firm:
ROTHWELL, Craig et al. (NZ)
Download PDF:
Claims:
Claims:

1. A system for indicating articles to a manual grader during a manual grading operation, the system comprising: an upstream inspection device configured to inspect a plurality of articles conveyed on a conveyor, wherein the upstream inspection device comprises an image capture device to capture an inspection image of the plurality of articles; a processor configured to detect one or more characteristics of the articles based on the inspection image; and a downstream indication device comprising a projection device configured to project a visual indication onto each article on the conveyor comprising the one or more characteristics to highlight each article comprising the one or more characteristics to a manual grader during manual grading of the plurality of articles.

2. The system as claimed in claim 1, wherein a characteristic of the articles is a defect and the indication device projects a visual indication onto each article comprising a defect to highlight each defective article to a manual grader.

3. The system as claimed in claim 1, wherein a characteristic of the articles is the absence of a defect and the indication device projects a visual indication onto each article without a defect to highlight each article without the defect to a manual grader.

4. The system as claimed in any one of claims 1 to 3, wherein the image capture device captures the inspection image in an upstream inspection area of the conveyor and the projection device projects a visual indication onto each article comprising the one or more characteristics in a manual grading area of the conveyor downstream of the inspection area.

5. The system as claimed in any one of claims 1 to 4, wherein the projection device projects a visual indication onto each article comprising the one or more characteristics at a time delay from when the image capture device captures the inspection image, and wherein the processor determines the time delay based on the conveyor speed and a distance between the image capture device and the projection device.

6. The system as claimed in claim 5, wherein the processor determines a coordinate for a detected characteristic or article with a detected characteristic based on the inspection image and the time delay is based on a distance between the coordinate and the projection device.

7. The system as claimed in claim 6, wherein the time delay is based on a distance between the coordinate and an upstream end of a manual grading area of the conveyor.

8. The system as claimed in any one of the proceeding claims, wherein the processor is configured to determine a coordinate for a detected characteristic or article with a detected characteristic, and update the coordinate based on a frame rate of the projection device and the speed of the conveyor to track the characteristic or article with a detected characteristic as it is conveyed through a manual grading area of the conveyor.

9. The system as claimed in any one of the proceeding claims, wherein the processor is configured to generate a projection image based on the inspection image, wherein the projection image comprises one or more indication areas, each indication area corresponding to a detected characteristic or article with a detected characteristic in the inspection image, and the projection device is configured to project the projection image onto the conveyor so that each indication area projects onto a characteristic or article with a detected characteristic.

10. The system as claimed in 9, wherein each indication area is an illuminated area within a remaining non-illuminated or contrasting area or background of the projection image.

11. The system as claimed in any one of claims 9 or 10, wherein the projection device projects the projection image onto the conveyor at a time delay from when the image capture device captures the inspection image, and wherein the processor determines the time delay based on the conveyor speed and a distance between the image capture device and the projection device.

12. The system as claimed in any one of claims 9 to 11, wherein the processor determines one or more coordinates for a detected characteristic or article with a detected characteristic based on the inspection image and generates the projection image using the coordinate(s) so that one or more coordinates for an indication area in the projection image corresponds to the coordinate(s) for the characteristic or article comprising a detected characteristic in the inspection image.

13. The system as claimed in 12, wherein the projection device projects the projection image onto the conveyor at a time delay from when the image capture device captures the inspection image, and wherein the processor determines the time delay based on the conveyor speed and a distance between the coordinate(s) and the projection device or an upstream end of a manual grading area and/or projection image.

14. The system as claimed in claim 12 or 13, wherein the processor updates the coordinate(s) for each indication area in the projection image based on a frame rate of the projection device and the speed of the conveyor so that each indication area tracks the characteristic or article with the detected characteristic as it is conveyed through a manual grading area of the conveyor.

15. The system as claimed in claim 14, wherein the processor updates the projection image regularly to provide a continuous visual indication to each detected characteristic or article with detected characteristic as it is conveyed through the manual grading area of the conveyor.

16. The system as claimed in any one of claims 9 to 15, wherein the image capture device captures a plurality of inspection images at an image capture device frame rate, and wherein the processor generates the projection image based on more than one inspection image.

17. The system as claimed in claim 16, wherein the projection image has a greater length in the conveyor direction than one said inspection image.

18. The system as claimed in any one of the proceeding claims, wherein: the image capture device comprises an area scan camera to provide a 2D inspection image; or the image capture device comprises a line scan camera configured to capture line scan images at a rate based on the speed of the conveyor, and wherein the processor is configured to stitch a plurality of line scan images together to provide a 2D inspection image.

19. The system as claimed in any one of claims 9 to 11, wherein the image capture device comprises a line scan camera configured to capture line scan images at a rate based on the speed of the conveyor, and wherein the processor is configured to: stitch a plurality of line scan images together to provide an uncropped inspection image that extends over a manual grading area of the conveyor, and generate the projection image, wherein the projection image is a cropped portion of the inspection image corresponding to the manual grading area of the conveyor.

20. The system as claimed in any one of the preceding claims, wherein the processor is configured to receive a signal indicative of a speed of the conveyor, and wherein where the processor makes a determination based on the speed of the conveyor, the processor makes said determined based on the signal.

21. The system as claimed in any one of the preceding claims, wherein the processor is in communication with the conveyor, and is configured to control a speed of the conveyor and/or stop and/or start the conveyor based on the detection of a characteristic or article with a detected characteristic.

22. The system as claimed in claim 21, wherein the processor is configured to slow down or stop the conveyor if the number of detected characteristics/ articles with detected characteristics is greater than a predetermined threshold.

23. The system as claimed in any one of the preceding claims, wherein the articles are fruit.

24. The system as claimed in claim 23, wherein the system is configured to indicate defective fruit to a grader during a manual fruit grading operation.

25. The system as claimed in claim 24, wherein the defect is fruit juice on a surface of a fruit.

26. The system as claimed in any one of claims 23 to 25, wherein the fruit is kiwifruit.

27. A method for indicating articles to a manual grader during a manual grading operation, the method comprising: capturing an inspection image of a plurality of articles on a conveyor, detecting one or more characteristics of the articles from the inspection image, and projecting a visual indication onto each article on the conveyor comprising the one or more characteristics to highlight each article comprising the one or more characteristics to a manual grader during manual grading of the plurality of articles

28. The method as claimed in claim 27, wherein the method comprises generating a projection image based on the inspection image, wherein the projection image comprise one or more indication areas, each indication area corresponding to a characteristic or article comprising the one or more characteristics in the inspection image, and the projection device is configured to project the projection image onto the conveyor so that each indication area projects onto an article comprising the one or more characteristics.

29. The method as claimed in claim 28, wherein each indication area is an illuminated area within a remaining non-illuminated or contrasting area or background of the projection image.

Description:
AN ARTICLE INDICATION SYSTEM

Field of Invention

The invention generally relates to the field of article conveying systems, article inspection systems and article sorting systems. More particularly, the invention relates to an article identification system to assist manual grading/sorting of articles in article conveying systems.

Background to the Invention

A common type of object that is graded and sorted during conveyance by a conveyor system is fruit. Fruit may be sorted based on criteria such as weight, shape, colour, ripeness and any other characteristic, such as surface blemishes or defects. Some conveyor systems comprise devices to measure these characteristics while the fruit is being transported. For example, the characteristics can be detected and measured by visual scanning devices positioned above the conveyor. The position of each fruit with its respective characteristics can be tracked so that a discharge mechanism causes items of fruit to be unloaded from the conveyor and sent to the required destination, for example towards a chute or onto another conveyor. Similar systems can be used to grade and sort other types of product based on similar characteristics or other characteristics specific to the type of product.

Many common fruit defects can be detected by existing fruit inspection means, which may be provided to a conveyor system. These existing inspection systems can detect and automatically grade fruit without requiring human intervention, and therefore offer labour efficiency gains for the processing of fruit.

However, some fruit grading and sorting operations require human involvement by manual graders. One such example is the manual grading of defective fruit resulting from burst overripe fruit. Overripe fruit (such as kiwifruit) can break down and can easily burst prior to or when being deposited onto a conveying system for inspection and sorting. An overripe burst fruit can release juice onto surfaces of the components of a conveyor system, such as conveyor rollers and belts and the like. The juice can transfer to other items of fruit being conveyed.

Fruit contaminated by juice is often considered to be defective because, if it is not removed, the juice results in a fungus called Alternaria. It is estimated that an average of 15 fruit is lost for every burst item of fruit. Any packed fruit that has juice on it will develop fungal growth and will need to be repacked.

Human graders may be employed to remove soft fruit and fruit contaminated by juice from a conveying system as soon as possible, prior to subsequent sorting and/or packing operations. If necessary, where a burst fruit or fruit juice is observed, the conveyor system may be paused to allow graders to wash or remove fruit juice from the conveyor system to prevent or reduce contamination of other fruit.

There is a significant cost associated with human grading or sorting of articles. Manually sorting articles can be a difficult task. High levels of attention must be maintained for an extended time period to observe defects or other characteristics. In a fruit sorting operation, due to the typically large volume of fruit being conveyed past a human grader, some defective fruit is missed, resulting in a percentage of defective fruit remaining for subsequent packing and dispatch. Such difficulties can be presented in any manual grading operation where large volumes of articles are conveyed past a manual grader.

Summary of the Invention

It is an object of the invention to provide a system for indicating articles to a manual grader in a manual grading operation that addresses one or more of the above-mentioned problems, and/or to provide a system for indicating articles to assist manual graders to correctly identify and select articles with or without defects or other characteristics, and/or to provide the related industry with a useful choice. According to a first aspect of the invention, the present invention provides a system for indicating articles to a manual grader during a manual grading operation. The system comprises an upstream inspection device configured to inspect a plurality of articles conveyed on a conveyor. The upstream inspection device comprises an image capture device to capture an inspection image of the plurality of article. A processor is configured to detect one or more characteristics of the articles based on (i.e. from) the inspection image. The system also comprises a downstream indication device comprising a projection device configured to project a visual indication onto each article on the conveyor comprising the one or more characteristics to highlight each article comprising the one or more characteristics to a manual grader during manual grading of the plurality of articles.

In some embodiments, a characteristic of the articles is a defect and the indication device projects a visual indication onto each article comprising a defect to highlight each defective article to a manual grader. In other embodiments, a characteristic of the articles is the absence of a defect and the indication device projects a visual indication onto each article without a defect to highlight each article without the defect to a manual grader.

In some embodiments, the image capture device captures the inspection image in an upstream inspection area of the conveyor and the projection device projects a visual indication onto each article comprising the one or more characteristics in a manual grading area of the conveyor downstream of the inspection area.

In some embodiments, the projection device projects a visual indication onto each article comprising the one or more characteristics at a time delay from when the image capture device captures the inspection image, and wherein the processor determines the time delay based on the conveyor speed and a distance between the image capture device and the projection device. In some embodiments, the processor determines a coordinate for a detected characteristic or article with a detected characteristic based on the inspection image and the time delay is based on a distance between the coordinate and the projection device. For example, the time delay is based on a distance between the coordinate and an upstream end of a manual grading area of the conveyor.

In some embodiments, the processor is configured to determine a coordinate for a detected characteristic or article with a detected characteristic, and update the coordinate based on a frame rate of the projection device and the speed of the conveyor to track the characteristic or article with a detected characteristic as it is conveyed through a manual grading area of the conveyor.

In some embodiments, the processor is configured to generate a projection image based on the inspection image, wherein the projection image comprises one or more indication areas, each indication area corresponding to a detected characteristic or article with a detected characteristic in the inspection image, and the projection device is configured to project the projection image onto the conveyor so that each indication area projects onto a characteristic or article with a detected characteristic.

In some embodiments, each indication area is an illuminated area within a remaining non- illuminated or contrasting area or background of the projection image.

In some embodiments, the projection device projects the projection image onto the conveyor at a time delay from when the image capture device captures the inspection image, and wherein the processor determines the time delay based on the conveyor speed and a distance between the image capture device and the projection device.

In some embodiments, the processor determines one or more coordinates for a detected characteristic or article with a detected characteristic based on the inspection image and generates the projection image using the coordinate(s) so that one or more coordinates for an indication area in the projection image corresponds to the coordinate(s) for the characteristic or article comprising a detected characteristic in the inspection image. In some embodiments, the projection device projects the projection image onto the conveyor at a time delay from when the image capture device captures the inspection image, and wherein the processor determines the time delay based on the conveyor speed and a distance between the coordinate(s) and the projection device or an upstream end of a manual grading area and/or projection image.

In some embodiments, the processor updates the coordinate(s) for each indication area in the projection image based on a frame rate of the projection device and the speed of the conveyor so that each indication area tracks the characteristic or article with the detected characteristic as it is conveyed through a manual grading area of the conveyor.

In some embodiments, the processor updates the projection image regularly to provide a continuous visual indication to each detected characteristic or article with detected characteristic as it is conveyed through the manual grading area of the conveyor.

In some embodiments, the image capture device captures a plurality of inspection images at an image capture device frame rate, and wherein the processor generates the projection image based on more than one inspection image.

In some embodiments, the projection image has a greater length in the conveyor direction than one said inspection image.

In some embodiments, the image capture device comprises an area scan camera to provide a 2D inspection image; or the image capture device comprises a line scan camera configured to capture line scan images at a rate based on the speed of the conveyor, and wherein the processor is configured to stitch a plurality of line scan images together to provide a 2D inspection image. In some embodiments, the image capture device comprises a line scan camera configured to capture line scan images at a rate based on the speed of the conveyor, and wherein the processor is configured to: stitch a plurality of line scan images together to provide an uncropped inspection image that extends over a manual grading area of the conveyor, and generate the projection image, wherein the projection image is a cropped portion of the inspection image corresponding to the manual grading area of the conveyor.

In some embodiments, the processor is configured to receive a signal indicative of a speed of the conveyor, and wherein where the processor makes a determination based on the speed of the conveyor, the processor makes said determined based on the signal.

In some embodiments, the processor is in communication with the conveyor, and is configured to control a speed of the conveyor and/or stop and/or start the conveyor based on the detection of a characteristic or article with a detected characteristic. For example, in some embodiments the processor is configured to slow down or stop the conveyor if the number of detected characteristics/ articles with detected characteristics is greater than a predetermined threshold.

In some embodiments, the articles are fruit. The system may be configured to indicate defective fruit to a grader during a manual fruit grading operation. In some embodiments the defect is fruit juice on a surface of a fruit. In some embodiments, the fruit is kiwifruit.

According to a second aspect, the present invention provides a method for indicating articles to a manual grader during a manual grading operation. The method comprises capturing an inspection image of a plurality of articles on a conveyor, and detecting one or more characteristics of the articles from the inspection image e.g. via image processing of the image, and projecting a visual indication onto each article on the conveyor comprising the one or more characteristics to highlight each article comprising the one or more characteristics to a manual grader during manual grading of the plurality of articles. Preferably the method comprises generating a projection image based on the inspection image, wherein the projection image comprise one or more indication areas, each indication area corresponding to a characteristic or article comprising the one or more characteristics in the inspection image, and the projection device is configured to project the projection image onto the conveyor so that each indication area projects onto an article comprising the one or more characteristics. For example, each indication area is an illuminated area within a remaining non-illuminated or contrasting area or background of the projection image.

The second aspect may include any one or more method steps as performed by the processor in the first aspect of the invention as described in any one or more of the above statements relating to the first aspect.

The term "camera" is used in this specification to refer to a device for capturing and/or sensing light, including visible and non-visible light. It will be understood that, in any of the embodiments of the invention described, other types of sensors may be used including, but not limited to: other types of light sensors; heat sensors; spectrometers; magnetometers; range sensors and noise sensors (e.g. microphones). These other types of sensors may be substituted for the camera in the described embodiments or may be used in addition to the described cameras.

Unless the context clearly requires otherwise, throughout the description and the claims, the words "comprise", "comprising", and the like, are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense, that is to say, in the sense of "including, but not limited to".

The entire disclosures of all applications, patents and publications cited above and below, if any, are herein incorporated by reference. Reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any form of suggestion that that prior art forms part of the common general knowledge in the field of endeavour in any country in the world.

The invention may also be said broadly to consist in the parts, elements and features referred to or indicated in the specification of the application, individually or collectively, in any or all combinations of two or more of said parts, elements or features.

Further aspects of the invention, which should be considered in all its novel aspects, will become apparent to those skilled in the art upon reading of the following description which provides at least one example of a practical application of the invention.

Brief Description of the Drawings

One or more embodiments of the invention will be described below by way of example only, and without intending to be limiting, with reference to the following drawings, in which:

Figure 1 illustrates a fruit defect detection and indication system to indicate to a manual sorter fruit that is defective.

Figure 2 is a schematic representation of one embodiment of a fruit defect detection and indication system according to the present invention.

Figure 3 is a schematic representation of another embodiment of a fruit defect detection and indication system according to the present invention.

Figure 4 is a schematic representation of another embodiment of a fruit defect detection and indication system according to the present invention.

Figure 5 is a flow diagram of method of inspecting, detecting and indicating defective fruit for manual grading by a person.

Figure 6 shows a projection image with indication areas. Brief Description of Preferred Embodiments of the Invention

Figure 1 shows a system 1 for indicating or highlighting articles of fruit to a manual grader 2 during a manual grading operation. The grader 2 is a person who is present alongside a conveyor 3, to inspect fruit 4 on the conveyor 3 and grade or sort the fruit based on whether an item of fruit is defective or has a defining characteristic. The fruit is stable on the conveyor (i.e. does not bounce around) and is arranged in a single layer on the conveyor for inspection. The conveyor may be a continuous belt-type conveyor however any suitable conveyor may be used and may include rollers or carriers or holders to singulate or maintain the relative positions of the fruit on the conveyor. A defective fruit is a fruit that has one or more defects such as surface blemishes or surface contamination or may be a fruit that is unacceptable based on its characteristics, such as size, form or colour or other defect such as ripeness (overripe and soft fruit). A preferred implementation for a system according to the present invention is in the sorting/separating of fruit contaminated by fruit juice that has oozed or exploded from an overripe fruit. By highlighting a defective fruit to the grader 2, the system 1 makes the grader's job of identifying defective fruit easier.

The system comprises an inspection device 5 and an indication device 10. The inspection device 5 is located upstream of the indication device 10 with respect to a conveying direction Y of the conveyor.

The inspection device 5 is configured to inspect fruit for defects or one or more other characteristics (herein 'defects') as the fruit is conveyed on the conveyor past the inspection device 5. The device 5 is shown located above the conveyor 3 to inspect fruit passing beneath the inspection device 5, however the inspection device 5 may be below or beside the conveyor 3 and may be in view of an inspection area of the conveyor directly or indirectly, for example via one or more mirrors or prisms.

With reference to Figure 2, the inspection device 5 comprises an image capture device 6 such as a camera to capture or generate an inspection image of a plurality of fruit in a capture or inspection area 7 of the inspection device on the conveyor. The inspection device is configured to capture the plurality of fruit in a single inspection image. The system 1 includes a processor 8 running image processing software. The processor is configured to determine defects on any fruit presented in the inspection image. The processor 8 may determine a coordinate, or plurality of coordinates, for a defect in the inspection image or a coordinate, or plurality of coordinates, for each fruit in the image determined/classified as being defective due to the presence of a defect on or of the fruit. The coordinate(s) may be with respect to the inspection image and/or with respect to the conveyor. The inspection device 5 preferably comprises an illumination source (not shown) to deliver illumination suitable for capturing high quality images, as known in the art (including illumination comprising radiation in visible/non-visible spectrums as required). Automatic fruit inspection systems for sorting/grading fruit utilising image processing of images of fruit or other articles to classify fruit are known. The inspection device 5 of a system 1 according to the present invention may utilise any such known inspection systems and devices to determine/detect a defect and/or determine a defect/no defect classification of the fruit.

The indication device 10 comprises a projection device 11 to project a visual indication onto each defect or defective fruit on the conveyor based on the inspection image and identification of a defect and/or classification of a fruit as being defective in the inspection image. The visual indication (12 in Figure 1) highlights a potentially defective fruit to the grader 2, so that the grader can select the fruit from the conveyor for further manual inspection and final determination of the defect status of the fruit. The projection device 11 projects the indication 12 onto each defect/defective item of fruit in a manual grading area of the conveyor downstream of the image capture device and inspection area of the conveyor in which the inspection image is captured.

The processor 8 determines where on the conveyor the projection device 10 is to project the visual indication 12 and/or when to project the visual indication 12 onto the conveyor to highlight a defect/defective fruit based on the inspection image. The processor 8 uses the line or conveyor speed to determine a location and/or when to project a visual indication to correspond to a defect/defective fruit on the conveyor.

In Figure 2, processor 8 is indicated as part of the inspection device. One or more processors may be provided to implement the various software tasks of the system and the one or more processors may be provided together with the inspection device or indication device or both. The term 'processor' used herein may refer to a single processor or a plurality of processors. In certain embodiments, processor 8, or one or more processors of a plurality of processors, may be a remote processor, which communicates with other components in system 1 via a network (for example a cellular network, or another network optionally comprising various configurations and protocols including the Internet, intranets, virtual private networks, wide area networks, local networks, private networks using communication protocols proprietary to one or more companies - whether wired or wireless, or a combination thereof).

The projection device 11 may be adapted to project a projection image (i.e. a 2-dimensional image) onto the conveyor. The processor 8 generates the projection image from the inspection image. The projection image 'maps' to the inspection image. The projection image may be a 'masked' image of the capture image. With reference to Figure 6, the projection image 14 comprises an indication or indication area 15 corresponding to a defect or a defective fruit identified/detected in the inspection image, and a contrasting 'background' or non-indication area 16 outside of the indication area. The indication area 15 may be an illuminated area such as a white light or bright illumination area, and the remaining area 16 of the projection image may be a dark or non-lit/non-illuminated area, or may be less illuminated than the illuminated area. The projection image may comprise one or more illumination/indication areas 15, each illumination/indication area corresponding to a defect or defective fruit determined by the inspection device 5. The processor is configured to generate, and the projection device is configured to project, a projection image comprising more than one indication or indication area 15 corresponding with more than one article comprising a defect. An overlay of the projection image with one or more inspection images may position each illumination/indication area in the projection image over a defect and/or defective fruit in the inspection image(s). The projection device 11 projects the projection image onto the conveyor after a time delay from when the inspection image was captured by the image capture device so that the illumination or indication area strikes the defect or defective fruit in the manual grading area IB of the conveyor 3. The inspection device 5 is located at a known distance from the indication device 5. The time delay may be calculated as the known distance between the image capture device 6 and the projection device 11 divided by the conveyor speed. The illumination or indication area is projected onto the fruit to provide the visual indication 12 to the manual grader. The projection device 11 may illuminate a portion of the item of fruit (for example a defect area of the fruit) or may illuminate substantially an entire area of the fruit visible to the projection device. The indication may be a projection of visible light onto the fruit. The light may be white light or other coloured light, may be circular or other shape, such as a cross shape. Alternatively, the indication may be a dark or non-lit area and areas outside the indication area may be lit/illuminated. The preferred form of indication may be able to be selected, and may be chosen based on the lighting conditions of the surrounding environment.

Preferably the projection device continues to provide an indication to the defect or defective fruit as the fruit travels along the conveyor in the manual grading area of the conveyor. The processor may be configured to generate/update the projection image regularly, for example many times per second, so that the indication in the projection image tracks or moves at a rate matched to the conveyor speed of the conveyor to effectively continually indicate the defect or defective fruit to the grader.

The processor 8 may determine a position or coordinate for the defect or defective fruit in the inspection image or projection image or both. The coordinate or position of the indication area in the projection image may be updated (for example many times per second) to move the position of the indication in speed with the conveyor. Where the coordinate of a defect in the capture image and/or a first projection image is (x, y), where the x coordinate is a distance in a direction orthogonal to the conveying direction from a datum point, and the y coordinate is a distance in the conveying direction from a datum point, then the coordinate in a second or subsequent/next projection image is equal to (x, y + d), where d is equal to the conveyor speed divided by the frame rate or refresh rate of the projection image. For example, if the processor generates/updates the projection image at 20 frames/images per second and the conveyor is travelling at lm per second, then the indication coordinate in the next projection image is (x, y + 50mm). The datum point may be a point in the capture/projection images, e.g. a corner of the respective image.

The process of capturing an inspection image and generating and projecting a projection image may be repeated regularly, for example many times per second. The inspection device preferably captures an inspection image many times per second. The projection image is also preferably generated/updated and projected many times per second. A projection image may be generated at the same frame rate as the image capture frame rate for the inspection image, or the projection image generation and/or projection rate may be different to the inspection image generation rate. The capture rate for the inspection image may be higher than the generation or projection rate for the projection image. For example, the fruit may rotate as they are conveyed through an inspection area of the inspection device. A defect may be present on one side of the fruit only, so a capture rate is preferably high enough to capture an image of all sides of the fruit as the fruit rolls on its axis as it is conveyed through the inspection area in order to detect defects anywhere on the fruit. A capture rate of 50fps may be sufficient for the image capture device to inspect all sides of the fruit in an inspection area length of around lm and conveyor speed of lm/s in a system in which a fruit rotates completely within a lm length of a conveyor, for example. The rate at which a fruit rolls on a conveyor may be dependent on the conveyor design. A person skilled in the art may calculate the speed of the conveyor or length of the inspection area/inspection image and the frame rate for the camera to ensure a fruit has rolled entirely around within the bounds of the inspection image to capture all defects within the image. The projection image generation and/or projection rate may be lower and is preferably sufficiently high to present a 'smooth' or visually continuous indication to the grader. A projection rate of around 20fps may present a visually continuous indication to a grader, for example. The processor may generate the projection image at the same rate as the capture rate of the image capture device. The projection device may project the projection image at a lower rate than the projection image generation rate.

The projection image may be based on more than one inspection image. The projection image may 'map' to more than one inspection image, i.e. the projection image may map to many overlaid inspection images. As a fruit is rotated in the inspection area and/or as new fruit enter the inspection area, new defects are detected in the inspection image and introduced to the next generated projection image. The projection image may have a greater length (in the conveyor direction) than the inspection image. Preferably the length of the projection image spans a length of a manual grading area of conveyor, such that upstream and downstream ends of the projection image projected onto the conveyor correspond to upstream and downstream ends of the manual grading area. The projection image continues to be updated/generated based on each new inspection image captured by the image capture device.

The processor may generate an uncropped projection image and a cropped projection image. The cropped projection image is cropped from the uncropped projection image and has a length shorter than the length of the uncropped image. The width of the uncropped and cropped images may also be different but, in some embodiments, is preferably the same. The projection device projects the cropped projection image onto the conveyor to project the indication(s) onto the defect(s) or defective fruit(s). The length of the cropped projection image preferably corresponds to a length of the manual grading area of the conveyor, and the length of the uncropped image extends upstream of the grading area. The uncropped projection image is generated from the inspection image as described above to include indications for the defects/defective fruit. The indications appear in the uncropped image and therefore appear in the cropped image as the fruit enters the manual grading area as the position of the fruit in the uncropped image is determined before the fruit enters the grading area.

In some embodiments, the processor may determine a location for a defect or defective fruit from the inspection image with respect to the conveyor. For example, a defect or defective fruit may be determined from the inspection image at an (x, y) coordinate of the conveyor. The coordinate with respect to the conveyor may be determined from a known position of the image capture device and the conveyor. The processor may then translate this coordinate to a downstream coordinate in the grading area based on a known distance D between the image capture device and the projection device to give a coordinate in the grading area of (x, y + D). The projection device may then project an indication to the defect or defective fruit when at the coordinate (x, y + D).

The coordinate (x, y) in the upstream inspection area is an upstream location and the coordinate (x, y + D) in the downstream grading area is a downstream location. The projection device projects the indication onto the downstream location at a time delay from when the upstream location was determined by the inspection device so that the indication strikes the defect or defective fruit when it is at the downstream location. The time delay may be calculated as the known distance between the image capture device and the projection device divided by the conveyor speed.

Preferably the projection device projects an indication onto a defect or defective fruit as the fruit enters the manual grading area. The processor may determine a coordinate for a defect or defective fruit from the inspection image with respect to the conveyor (the upstream location). The coordinate with respect to the conveyor may be determined from a known position of the image capture device and the conveyor. The processor may determine a downstream location for the defect/defective device at a known upstream end of the manual grading area, this location being the 'x' or width position of the defect/defective fruit and the known upstream end of the grading area.

The projection device projects the indication onto the conveyor at the downstream location at a time delay from when the inspection image was captured by the image capture device so that the illumination or indication area strikes the defect or defective fruit as the fruit enters the manual grading area of the conveyor. The time delay may be calculated as the distance between the upstream location/coordinate for the defect/defective fruit determined by the inspection device and a known location of the start of the grading area divided by the conveyor speed. The projection device, and the indication projected thereby, preferably continues to 'track' the fruit as is it is conveyed through the grading area, based on the frame/projection rate for the projection device and the speed of the conveyor, as described above, i.e. where the conveyor speed is lm/s and the projection device projects an indication at a rate of 20 times per minute, the coordinate is updated from (x, y) to (x, y + 50mm) every 50milliseconds as the fruit is conveyed through the grading area of the conveyor.

Where the processor makes a determination based on the speed of the conveyor, the processor may utilise a predetermined or known conveyor speed or may receive as an input from a sensor a signal indicative of an actual speed of the conveyor. Where the processor is in communication with a conveyor speed sensor, the system is preferably configured to use the sensed speed of the conveyor in the determination of where and/or when to project the indication to the defect/defective fruit, to avoid incorrectly identifying defects/defective fruit to the grader due to changes in conveyor speed. In some embodiments, the system 1 may be configured to interface with the conveyor to control the conveyor speed and/or stop and/or start the conveyor. For example, the system may slow down or stop the conveyor if the number of defects/defective fruit detected is greater than a predetermined threshold. This slowing down or stopping provides the manual grader with sufficient time to manually grade the fruit. The processor may be configured to restart the conveyor after a time period has elapsed to maintain a desired throughput. If a conveyor is stopped for too long, downstream buffers of fruit can empty which reduces the overall throughput of a grading line. By controlling the conveyor speed and/or stopping and/or starting the conveyor, the system 1 can ensure defect fruit is removed from the conveyor whilst also maintaining a desired throughput.

The inspection image capture device 6 may comprise any suitable sensor on which an inspection system can determine a characteristic/defect of an article. As described above, the image capture device may comprise a camera sensor, such as a CMOS sensor or other known image capture technology. The capture device 6 may comprise an area scan camera/sensor, as indicated in Figure 2, to capture multiple fruit in a single image over a capture or inspection area 7 of the conveyor. The area scan camera may capture an image with a width/side dimension that corresponds to the width of the conveyor, or the image may extend a distance greater than the conveyor width or may extend for a portion of the width, although this is less preferred. The capture device may comprise more than one sensor/camera, for example two cameras each capturing a portion (e.g. half) of the width of the conveyor.

Alternatively, the image capture device 6 may comprise a line scan camera/sensor as indicated in Figure 3, each image being one pixel wide and with a length extending across the conveyor (e.g. orthogonal to the conveying direction). The length of the line scan image may correspond to the width of the conveyor, or may be greater than the width, or may extend for a portion of the width, although this is less preferred. The image capture device may comprise more than one line-scan camera, for example two cameras each capturing a portion (e.g. half) of the width of the conveyor. The line scan camera may capture images at a rate based on the speed of the conveyor so that many line scan images may be stitched together side-by-side to produce an area image equivalent to an 'area scan image' of the inspection area. The line scan images are combined together in a 'scrolling mosaic', such that an earliest line scan image 'drops off' a downstream end of the area scan as a latest line scan image is added to an upstream end of the area scan, to generate the 2D area scan image. Other than the method for generating the inspection image, the embodiments of Figures 2 and 3 are implemented in the same way.

A line scan camera embodiment may provide a lower cost system, and line scan cameras typically provide higher resolution images than area scan cameras. However, an area scan camera may be preferred in order to capture an image of all sides of an item of fruit as it rolls as it is conveyed through the inspection area to ensure no defects are missed. Stitching together line scan images may miss a defect on an obscured side of an item of fruit, as the fruit may not roll completely around its axis within a length provided by a single pixel line provided by the line scan camera.

Figure 4 illustrates a further embodiment utilising a line scan camera. In this embodiment, the line scan camera captures images at a rate based on the speed of the conveyor so that many line scan images may be stitched together side-by-side. A sufficient number of line-scan images are stitched together to produce an uncropped inspection image that extends through the manual grading area. As the line scan images are captured at a rate matched to the speed of the conveyor, the uncropped inspection image maps directly to the position of fruit on the conveyor in the manual grading area of the conveyor. The projection device projects a projection image that is a cropped from the inspection image, i.e. the projection image is a cropped portion of the inspection image corresponding to the manual grading area of the conveyor. In this embodiment, the location or coordinates of the defects/defective fruit detected in the inspection image need not be calculated since the projection image is a portion of the inspection image. The processor may generate the projection image by cropping the projection image from the inspection image and additionally modifying the projection image to present a 'masked' image of the capture image, whereby the projection image comprises an indication or indication area corresponding to a defect or a defective fruit identified in the inspection image, and a contrasting 'background' or non-indication area outside of the indication area, as described above.

In the above described embodiments, the projection device is adapted to project a 2D image, for example the projection device is a projector adapted to project an image comprising many pixels. Each visual indication projected to a defect or defective fruit is an area or group of pixels within the 2D image. An example projection device is an EPSON 2000Lu WXGA Laser Projector. However, the present invention may be implemented with other projection devices, such as a directional bulb, LED and/or laser light source that is adapted to provide a narrow beam of illumination in a controllable direction to project an indication onto the defect or defective fruit. For example, a projection device may comprise a light source with lens, and wherein the light source and/or the lens is/are movably mounted and controlled via one or more actuators to direct a focused illumination to a defect or defective fruit. The processor may be configured to drive the actuators to position the light source and/or lens to project the indication to a defect or defective fruit based on one or more coordinates for the defect or defective fruit. The indication device may comprise a plurality of such projection devices, to be configured to indicate a plurality of defects/defective fruit to a grader in the manual grading area of the conveyor.

A method for indicating a defective fruit to a manual grader during a manual grading operation is illustrated in Figure 5. The method of Figure 5 generally reflects a method employed in each of the systems of Figures 2 to 4, as described above. Incoming fruit 4 is transported on a conveyor to the inspection device 5. The inspection device performs defect detection, by generating an inspection image of the fruit on the conveyor and detecting a defect or defective fruit from the inspection image. A projection image is built or generated based on the inspection image and defect detection from the inspection image. The projection image includes indication areas as described above, each indication area of the projection image corresponding to a detected defect or defective fruit. Preferably each indication area is an illuminated area of the projection image. The projection image is projected onto the conveyor at a time delay after the inspection image was captured in which the defect or defective fruit was detected. The time delay is based on a distance between the image capture device and the image projection device and the line speed of the conveyor. By projecting the projection image onto the conveyor, the illuminated area of the projection image is projected onto the fruit, to indicate a potentially defective fruit to the grader. The grader makes a manual grading of the indicated fruit and makes a final determination of whether the item of fruit is a good fruit or a reject fruit. As described above, in some embodiments, the projection device may instead project a visual indication other than in a projection image, such as a beam of light directed to a coordinate corresponding to a defect or defective fruit.

The invention has been described with reference to a system for indicating defective fruit to manual graders to assist in a manual fruit grading operation. However, one skilled in the art will appreciate that the invention may be implemented in a system to assist manual grading of articles other than fruit, and therefore should not be limited to a system for use in the grading of fruit. A system according the present invention may be utilised in the manual grading of other food products, such as meat, poultry or fish products, or plant/plant based food products, processed food products, and non-food products/articles such as mechanical or electrical components. Furthermore, the system may be used to grade more than one type of article at the same time. For example, apples, pears and kiwi fruit may be graded at the same time. The processor 8 may determine from the inspection image the type of fruit, for example based on size, shape, colour or other characteristic. Defects or other characteristics may be assessed based on the type of article. Visual indications may also be provided based on the type of article. For example, a different colour or visual indication may be projected to the articles based on the type of article.

Furthermore, the invention may be used to indicate articles comprising one or more characteristics other than defects. For example, the invention may be used to indicate articles such as fruit that have one or more desirable or superior characteristics, or articles that are without defects, as opposed to indicating defective or unacceptable articles. There may be situations where there are many more articles with defects than without, such that manual grading may be better assisted/more efficient by indicating to a manual grader articles without defects for selection and subsequent packing and distribution. The absence of a defect may be considered a characteristic of the article. The system may be configurable to select whether the system indicates defective or non-defective articles, for example depending on an actual or anticipated defect/no defect rate. Thus, in the preceding description, one skilled in the art will appreciate that the term 'defect' may be replaced with 'characteristic' or 'quality characteristic' and the terms 'defective article' or 'defective fruit' may be replaced with 'non-defective article' or 'non-defective fruit'.

In some embodiments, a system according to the invention may indicate more than one class or grade of article to a manual grader. For example, the system may indicate a high, medium and low grade of article to a manual grader. Each visual indication projected onto a fruit or article may comprise a code and/or colour indicative of a grade from a plurality of possible grades. For example, three grades may be represented by the colours red, orange and green.

In a further example embodiment, a system according to the invention may additionally include an automatic discharge mechanism configured to automatically discharge an article from the conveyor based on one or more characteristics of an article as detected by the inspection device 5. In such an embodiment, the automatic discharge mechanism is located downstream of the inspection device 5. The automatic discharge mechanism may be located upstream of the indication device 10 and/or the manual grading area IB of the conveyor or may be located downstream of the indication device 10 and/or the manual grading area 13. The automatic discharge mechanism may be configured to discharge an article based on one or more first characteristics, and the indication device 10 may be configured to project a visual indication onto an article based one or more second characteristics different to the one or more first characteristics.

Where in the foregoing description reference has been made to integers or components having known equivalents thereof, those integers are herein incorporated as if individually set forth.

It should be noted that various changes and modifications to the presently preferred embodiments described herein will be apparent to those skilled in the art. Such changes and modifications may be made without departing from the spirit and scope of the invention and without diminishing its attendant advantages. It is therefore intended that such changes and modifications be included within the present invention. Also, the various embodiments described above may be implemented in conjunction with other embodiments, e.g., aspects of one embodiment may be combined with aspects of another embodiment to realize yet other embodiments. Further, each independent feature or component of any given assembly may constitute an additional embodiment.