AU2016102045A4 - A method and system of automating the detection of fresh cuts on an animal carcass - Google Patents

Abstract

The present invention relates to a method and system for the automating the detection of fresh cuts on an animal carcass being proceeded for its flesh. The invention has particular application to the removal of the pelt from a sheep carcass although it will be appreciated that this is not meant to be limiting. The method involves the steps of: a) defining a processing area which encompasses at least part of a surface of the carcass which has been mechanically worked by a cutting implement; f) detecting the thermal output from that surface; g) correlating differences in thermal output from the area to identify a fresh cut on the surface; h) taking a three-dimensional image of the processing area; and i) mapping the co-ordinates of the identified fresh cut. The co-ordinates can be used with an industrial robot equipped with tooling, such as clamping or gripping members, so that the removal of the pelt from the animal carcass (or other actions) may be automated. 20021 202 214 2/s /v IJI 212 212 102)20 200 216 202 214

Description

A METHOD AND SYSTEM OF AUTOMATING THE DETECTION OF FRESH CUTS ON AN ANIMAL CARCASS 2016102045 28 Nov 2016

TECHNICAL FIELD

The present invention relates to a system and method for the automating the detection of cuts on an animal carcass being proceeded for its flesh. The invention has particular application to the removal of the pelt from a sheep carcass although it will be appreciated that this is not meant to be limiting.

BACKGROUND ART

There is a growing trend in abattoirs for the automation of various parts of the slaughtering and processing of an animal carcass. This is primarily due to improving the safety and efficiency of the operation of the abattoir. A worker involved in the processing of a carcass may have to work with a variety of cutting implements, all of which have the potential for significant injury. The work is highly repetitive but requires relatively intense concentration.

Accordingly, the worker can become very fatigued in the course of performing their shift. It is well known that mistakes are more easily made when a worker is tired. These mistakes can result in injury to themselves or others or in the wastage of commercially valuable carcasses.

Some stages of the slaughtering and processing of an animal carcass are easier to automate than others. However, removing of the pelt from the carcass, the initial step in its "dressing", is one stage that is particularly problematic.

The removal of the pelt from an animal carcass is a relatively manual process. In particular, while tools exist for the gripping of the pelt, it is a relatively difficult process to automate.

This is because of the need to distinguish the pelt from the underlying flesh. Colour variation in a pelt can be significant, ranging from pristine off-white to dark brown and black, depending on the age, breed 1 and/or species of animal being processed. 2016102045 28 Nov 2016

Additionally, the presence of blood, exposed flesh, and fatty tissue can also greatly vary, presenting challenges to the use of equipment such as cameras that may be employed in an attempt to automate this process.

For this reason, human input continues to be heavily engaged in the removal of the pelt from the carcass, despite their disadvantages.

It is an object of the present invention to address the foregoing problems or at least to provide the public with a useful choice.

Further aspects and advantages of the present invention will become apparent from the ensuing description which is given by way of example only.

All references, including any patents or patent applications cited in this specification are hereby incorporated by reference. No admission is made that any reference constitutes prior art. The discussion of the references states what their authors assert, and the applicants reserve the right to challenge the accuracy and pertinency of the cited documents. It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that any of these documents form part of the common general knowledge in the art, in New Zealand or in any other country.

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".

DISCLOSURE OF THE INVENTION

According to one aspect of the present invention, there is provided a method of detecting a fresh cut on an animal carcass, the method including the steps of: 2 a) defining a processing area which encompasses at least part of a surface of the carcass which has been mechanically worked by a cutting implement; 2016102045 28 Nov 2016 b) detecting the thermal output from that surface; c) correlating differences in thermal output from the area to identify a fresh cut on the surface; d) taking a three-dimensional image of the processing area; and e) mapping the co-ordinates of the identified fresh cut.

According to another aspect of the present invention there is provided a system for detecting a fresh cut on an animal carcass, the system including: a camera configured to detect thermal output; a processor linked or otherwise connected to the camera, a three-dimensional imaging system communicative with the processor, and wherein the processor is configured to execute the following steps: a) defining a processing area which encompasses at least part of a surface of the carcass which has been mechanically worked by a cutting implement; b) detecting the thermal output from that surface with the camera; c) correlating differences in thermal output from the area to identify a fresh cut on the surface; d) taking a three-dimensional image of the processing area; and e) mapping the co-ordinates of the identified fresh cut.

According to another aspect of the present invention there is provided an article of manufacture having computer storage medium storing computer readable program code executable by a computer to 3 implement a method for detecting a fresh cut on an animal carcass, the code including: 2016102045 28 Nov 2016 computer readable program code defining a processing area which encompasses at least part of a surface of a carcass which has been mechanically worked by a cutting implement; computer readable program code detecting the thermal output from that surface with a camera configured to detect thermal output and communicative with the computer; and computer readable program code correlating differences in thermal output from the area to identify a fresh cut on the surface.

According to another aspect of the present invention there is provided an article of manufacture having computer storage medium storing computer readable program code executable by a computer to implement a method for detecting a fresh cut on an animal carcass, the code substantially as described above and including: computer readable program code taking a three-dimensional image of the processing area three-dimensional imaging system communicative with the computer; computer readable program code mapping the co-ordinates of the identified fresh cut.

The inventive method and apparatus is intended to better automate the process of certain mechanical actions involved in the processing of an animal carcass, for example, cutting or gripping the pelt of an animal carcass. The use of cameras that use thermal output, and in particular passive infrared radiation, to detect differences in surface temperature by measuring the heat emanating from the carcass being processed, allows fresh cuts and therefore the location of an edge of the pelt, to be identified. When used with a three-dimensional imaging system, the co-ordinates of the edge of the pelt can be sent to a robotic arm equipped with a clamping mechanism to automate its removal.

The animal carcass to be processed may be of any mammalian species that has flesh or pelt that may be commercially valuable. 4

In preferred embodiments of the invention, the animal carcass is that of a sheep. Reference shall now 2016102045 28 Nov 2016 be made throughout the remainder of this specification to the carcass being that of a sheep. However, it should be understood that this is not meant to be limiting and the method and apparatus of the present invention may also be readily applied to other animals, such as caprine or bovine species, by way of example.

The present invention is to be deployed at a carcass processing facility, at which sheep are humanely killed at or shortly after arrival (some facilities will hold the animals for at least several hours prior to killing so that they may recover from the stress of being transported), and their carcasses prepared for processing into cuts of meat for consumption.

During the initial stages at least, the carcass is suspended from an overhead rig by its hocks to better facilitate its processing.

Typically, the processing of the carcass is performed at a number of sequential stations, each equipped with appropriate tooling. The rig bearing the carcass is conducted to each station via an overhead rail or cable system.

The carcass of the sheep will be understood to be covered by a pelt or hide. In preferred embodiments, the invention is used to automate the removal of the pelt which needs to be removed in order to further process the carcass for its flesh. This action will be performed at a pelt removal station.

Thus it will be appreciated that in preferred embodiments of the present invention, the system and its associated apparatus may be deployed at or prior to arrival at the pelt removal station and reference to this effect shall now be made throughout the remainder of the present specification.

However, it is envisaged that there are other steps to the processing of the carcass to which the invention may be applied. For example, the invention may be used to detect the transition between the flesh of the carcass and its viscera and facilitate the removal of the latter portion of the carcass. In this case, the tooling at the relevant station at which this action may be performed could include cutting 5 equipment or the like. Persons skilled in the art will appreciate other parts of the process of working an 2016102045 28 Nov 2016 animal carcass that may be suitable for use with the present invention.

The pelt removal station will be understood to include an industrial robot with one or more robotic arms to which tooling, such as a clamping mechanism, is mounted.

Industrial robots are well known for use in automated processing. The industrial robot may be, without limitation: an articulated robot, a SCARA robot, a Delta robot, or a Cartesian coordinate robot. Therefore industrial robots with the necessary articulation to achieve the removal of the pelt are readily known and will be straightforward for a person skilled in the art to integrate into the pelt removal station. It will be appreciated that the industrial robot and/or arms of same include sensors and other ancillary equipment to provide positional feedback.

An example of a suitable clamping mechanism that may be readily employed in the present invention is described in the applicant's New Zealand Patent No. 631453 but it will be understood that this is not meant to be limiting.

Reference will now be made throughout the remainder of the specification to the tooling mounted to the robotic arms as being a clamping or gripping mechanism. However, this is not meant to be limited as will be recognised by persons skilled in the art.

To perform the present invention, the carcass must be mechanically worked prior to presentation to the processing station. The term should be understood to include, but not be limited to, the cutting of pelt and its partial separation from the underlying connective tissue.

One or more initial cuts, as part of the preparation process for the processing of the animal carcass, may have been performed to expose at least a portion of the underlying flesh.

This initial cut may be performed by a human worker at the carcass processing facility but may have also been performed by an industrial robot as part of an automated process. 6

Preferably, the portion of the carcass that has been mechanically worked is the brisket and/or 2016102045 28 Nov 2016 shoulders. In some embodiments, a portion or all of the circumference of the hocks may have been cut and the pelt on the legs stripped. However, this is not meant to be limiting; other portions may have been mechanically worked.

It will be understood that the processing area includes at least part of the cut portion of the carcass.

It will be appreciated that the carcass has a thermal output, the variation of which depends on the extent to which the underlying flesh of the carcass has been exposed. The pelt has an insulative effect, and will be cooler than the underlying flesh. Any flesh that is freshly exposed, such as when the pelt has been cut, will have a higher temperature than the pelt itself. This then allows the edge of the pelt to be accurately identified and an automated gripping tool directed to the site accordingly.

The inventors have found that using differences in the thermal output of a partially worked sheep carcass a more accurate method for identifying the location of the pelt than, for example, the use of colour photography or of three-dimensional data (or a combination of both).

The natural variances in the colours of the pelt and underlying flesh (which may be comprised of muscle, fat and connective tissue) compromise the easy detection of the edge of the pelt, making it difficult for automated positioning of tooling (such as the clamping mechanisms).

The use of three-dimensional data presents its own difficulties, particularly with respect to sheep, due to the woolly nature of the pelt. This can affect the profile of the carcass as it is positioned relative to the camera, again making it difficult for automated tooling to be positioned in such a way to further process the pelt.

This temperature difference may be detected through the use of a thermographic camera, such as a passive infrared (PIR) camera. This should be understood to be a camera that is capable of capturing images of objects which have a thermal output, and in particular, infrared radiation. Examples of PIR cameras suitable for use with the present invention are manufactured by the likes of Optris™ and 7

Xenics™ although persons skilled in the art will be aware this is not meant to be limiting. 2016102045 28 Nov 2016

The camera takes a thermographic image of the carcass, including the cut portion, and in preferred embodiments of the invention transmits this directly (or indirectly) to a central processing system that includes a processor for processing and analysis.

Of course, in some embodiments of the invention, the PIR camera may have sufficient processing capabilities to perform the necessary analysis so this is not meant to be limiting.

For a firmware and/or software (also known as a computer program) implementation, the techniques of the present disclosure may be implemented as instructions (for example, procedures, functions, and so on) that perform the functions described. It should be appreciated that the present disclosure is not described with reference to any particular programming languages, and that a variety of programming languages could be used to implement the present invention. The firmware and/or software codes may be stored in a memory, or embodied in any other processor readable medium, and executed by a processor or processors. The memory may be implemented within the processor or external to the processor. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, for example, a combination of a digital signal processor (DSP) and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. The processors may function in conjunction with servers and network connections as known in the art.

The steps of a method, process, or algorithm described in connection with the present disclosure may be embodied directly in hardware, in a software module executed by one or more processors, or in a combination of the two. The various steps or acts in a method or process may be performed in the order shown, or may be performed in another order. Additionally, one or more process or method 8 steps may be omitted or one or more process or method steps may be added to the methods and 2016102045 28 Nov 2016 processes. An additional step, block, or action may be added in the beginning, end, or intervening existing elements of the methods and processes.

The processor of the central processing system may be linked to, or otherwise in communication with, the robotic arm of the industrial robot that bears the clamping mechanism of the pelt removal station.

By analysing the thermographic image captured by the camera, the thermal variation allows the automation of the detection of the edge of the freshly cut pelt.

The thermographic image may require a number of processing steps to be performed before the transition between the cut edge of the pelt and the exposed flesh can be readily identified.

These processing steps can be performed by the processor using software identified as being appropriate for the task by a person skilled in the art. The inventors used a combination of Matlab™ and LabView™ software for processing the data and visualisation of same once processed but these are examples only and are not meant to be limiting.

Processing steps may include, but are not limited to, defining a region of interest of the carcass the subject of the thermographic image. The region of interest will be understood to include the portion of the carcass which has been mechanically worked.

This may include setting a threshold temperature to assist in filtering extraneous data from the thermographic image and allow better definition of the carcass portion. For example, the threshold temperature may be set at 25°C; any portion of the image at a lower temperature is appropriately filtered. This may be required to remove any equipment or personnel present in the processing station and which may have been captured in the thermographic image. 9

Another processing step may be the segmentation of the region of interest of the thermographic image 2016102045 28 Nov 2016 into individual blocks. This can enable further processing to identify specific points of interest on the cut portion of the carcass.

In preferred embodiments of the invention, these points of interest are the transition areas between the flesh and the pelt, specifically the cut edges of the pelt. It will be apparent to a person skilled in the art that there are a number of image processing techniques and algorithms that would be suitable for identifying the specific points of interest.

In preferred embodiments of the invention, cluster analysis (using k-means) is performed. This technique divides the data obtained from the region of interest into two clusters or groups. One cluster is associated with the flesh while the other is associated with the pelt.

This helps in the defining of the temperature difference between the pelt and the flesh and therefore the distinction between these portions of the carcass. These represent potential gripping points for the tooling of the processing station.

However, this is not meant to be limiting and other algorithms suitable for use in differentiating the pelt of a carcass from the underlying flesh would be readily apparent to a person skilled in the art.

Signals can then be sent by the central processing system to the clamping mechanism with the appropriate instructions to articulate it appropriately, relative to the gripping points that have been identified with the present method on the cut portion of the pelt.

The pelt can then be removed from the carcass without the need for human intervention.

It should be appreciated that the PIR camera effectively sees in two dimensions (X-Y, i.e. height and width). This may be sufficient for simple scenarios when the carcass is positioned in a default location in fixed and clearly defined area with known co-ordinates to which the arms of the industrial robot may be directed. However, a carcass is a relatively complex shape and furthermore an added layer of 10 complexity may be presented by its proximity to the PIR camera (a third dimension, Z- i.e, depth 2016102045 28 Nov 2016 dimension).

Therefore, preferred embodiments of the invention may include additional equipment to supplement the information provided by the PIR camera and provide even greater accuracy for automation of the removal of the pelt.

For example, preferred embodiments of the invention including three-dimensional imagery functionality. This can be useful for more accurate detection of the transition between the pelt and the flesh of the carcass.

For example, the use of three-dimensional imagery allows for the detection of variations in depth across the front (i.e. the brisket, shoulders and neck) of the carcass. Variations in depth may be an indication of start and end points of the exposed flesh relative to the pelt.

The three-dimensional imagery also provides information in the form of X-Y-Z co-ordinates to the robotic arms to which the clamping mechanism is fitted.

In these embodiments, the invention includes a three-dimensional imagery system which is communicative with the central processing system.

Such an imagery system may include equipment such as stereo cameras that are linked to the central processing station. Flowever, it will be appreciated that other techniques for three-dimensional imaging may be employed, for example, time-of-flight or structured light cameras mounted in front of the rig bearing the carcass. There may be a compromise in the quality of the data obtained via this technique.

In trials, the inventors used a Kinect™, which uses structured infrared, which has a different frequency range to that of passive IR. However, this is not meant to be limiting and as noted above, other equipment may be used.

It will be appreciated that the images taken with the three-dimensional imagery system need to be 11 mapped to the thermographic image. 2016102045 28 Nov 2016

Techniques for achieving this will be apparent to persons skilled in the art. For example, common points of interest to both images may need to be identified to assist in this mapping. These common points may include portions of the carcass and/or surrounding equipment (which may be fixed in place and thus have known co-ordinates to assist in calibration and mapping). A calibration rig may be also used to calibrate the images obtained with the PIR camera and the three-dimensional camera. The rig consists of a series of points positioned in space on a framework. These can be black on a light background, which can be captured by both the PIR and three-dimensional cameras. When used with reference markers on the robot, these can be used to determine the relative positions of these points in space. A transform matrix is applied to the data which can then be used on the data collected from the carcasses. This information can then be delivered to the clamping mechanism as co-ordinates.

In summary, the present invention offers a number of advantages including: • automated detection of cuts on animal carcasses that have been mechanically worked; • allows full or partial automation of the removal of the pelt of an animal carcass with a relatively high degree of accuracy; • reduces the number of human workers that need to employed in the strenuous and fatiguing task of the removal of the pelt of an animal carcass.

At the very least, the present invention offers the public a useful choice.

BRIEF DESCRIPTION OF DRAWINGS

Further aspects of the present invention will become apparent from the following description which is given by way of example only and with reference to the accompanying drawings in which: 12 is a flow chart of the steps of the process for detecting a fresh cut on a carcass;

Figure 1 2016102045 28 Nov 2016

Figure 2 is a schematic of the system used in the process of Figure 1;

Figures 3a & 3b are representative thermal images of carcasses captured by the passive infrared camera of the system of Figure 2;

Figures 4a & 4b are the thermal images of Figures 3a and 3b following implementation of a three-cluster algorithm;

Figure 5 is a thermal image of a carcass showing gripping points on the pelt identified with the present system and method.

BEST MODES FOR CARRYING OUT THE INVENTION

An exemplary system for detecting a fresh cut on an ovine (sheep) carcass is illustrated in Figure 1 as part of the overall process (generally identified by arrow 100) of slaughtering sheep for meat.

Sheep (102) are transported via a truck (104) to a processing facility, such as an attaboir (106), for slaughter. This is typically achieved humanely by stunning the sheep and cutting its throat to allow it to bleed out. Following its slaughter, the carcass of an individual sheep (102) is positioned on a rig (108) by suspending it by its hocks (110) and transported to an initial processing station (110).

Flere the carcass (102) is prepared by an appropriate cutting action on the pelt (114) about its brisket and/or around its shoulders to expose some of the underlying flesh (116).

The rig is then delivered to a pelt removal station (118) so that the pelt may be removed from the carcass.

Turning now to Figure 2, the pelt removal station (118) includes a system (200) for detecting a fresh cut on the carcass (102) and this shall now be described. 13

The system 2016102045 28 Nov 2016

The system (200) for detecting a fresh cut on the carcass (102) includes a controller (202), a passive infra-red (PIR) camera (204), a three-dimensional imaging system (206), and an industrial robot (208), to which a clamping mechanism (210) is mounted on a pair of arms (212).

The controller (202) has a processor (214), memory (216), and other components typically present in such computing devices.

In the exemplary embodiment illustrated the memory (216) stores information accessible by processor (214), the information including instructions (218) that may be executed by the processor and data (220) that may be retrieved, manipulated or stored by the processor. The memory may be of any suitable means known in the art, capable of storing information in a manner accessible by the processor, including a computer-readable medium, or other medium that stores data that may be read with the aid of an electronic device.

The processor (214) may be any suitable device known to a person skilled in the art. Although the processor and memory (216) are illustrated as being within a single unit, it should be appreciated that this is not intended to be limiting, and that the functionality of each as herein described may be performed by multiple processors and memories, that may or may not be remote from each other.

The instructions (218) may include any set of instructions suitable for execution by the processor (214). For example, the instructions may be stored as computer code on the computer-readable medium. The instructions may be stored in any suitable computer language or format.

Data (220) may be retrieved, stored or modified by processor (214) in accordance with the instructions (218). The data may also be formatted in any suitable computer readable format. Again, while the data is illustrated as being contained at a single location, it should be appreciated that this is not intended to be limiting - the data may be stored in multiple memories or locations. 14

The data (220) may include a record (222) of control routines for the system (200), including control 2016102045 28 Nov 2016 routines for individual components of the system such as the robotic arms (212). Further, the data may include data obtained from various sensors of the system, including the passive infra-red (PIR) camera (204), and the three-dimensional imaging system (206).

During some experimental trials, the inventors used a Xenics™ Gobi-640- GigE PIR camera (http://www.xenics.com/en/gobi-640-gige) while in others, an Optris™ PI400 thermal camera (http://www.optris.com/thermal-imager-pi400), was used. However, these are just examples, and a person skilled in the art will readily utilise other brands and types of PIR cameras for use in the present invention.

Whatever PIR camera (204) is used, it is linked to the controller (202) which includes the processor (214) required to carry out the processing and analysis for detecting the fresh cut on the carcass (102).

As noted previously, the system (200) also includes a three-dimensional imaging system (206) that is linked to the controller (202). The controller in turn is communicative with the pair of robotic arms (212) of the industrial robot (208), to which clamping members (210) are mounted.

In use, the carcass, which has been partially worked to expose the underlying flesh (116) as a pre-curser step to the removal of the pelt (114), approaches the PIR camera (204). Typically the carcass is suspended from the rig (112) to present the neck, shoulders and front of the chest area to the PIR camera.

The PIR camera (204) has a field of vision that substantially corresponds to at least a portion of the carcass, including the area that has been cut to expose the flesh (116).

It will be appreciated that the pelt (114) has an insulative effect on the carcass (102); any flesh freshly exposed by a cut will have a higher thermal output than the pelt itself. Thus, there is a temperature transition at the area at the edge of the cut. In testing, the inventors have found that the difference in temperature between the pelt and the flesh is around 3 to 4°C. 15

In practice, this differential may be less as there was a five to ten minute delay between slaughter of 2016102045 28 Nov 2016 the animal and the time testing was performed. Typically, the carcass (102) will arrive at the pelt removal station (118) between 90 seconds and three minutes after slaughter.

This difference is detected by the PIR camera (204) which takes a thermal image (224) and transmits it (226) to the controller (202). Examples of thermal images are shown in Figures 3a and 3b. These show two separate sheep carcasses from which the pelt has partially been removed from the front leg and brisket region.

Image analysis and processing

Returning now to Figure 2, the image (224) of the carcass (102) may require a number of processing steps to be performed before the transition between the cut edge of the pelt (114) and the flesh (116) is identified such that the clamping mechanisms (210) of the robotic arms (212) can then be automatically directed to it.

Image processing was performed by using a combination of Matlab and LabView software (Matlab™ for processing the data, LabView™ for visualisation of the processed data).

The image of the carcass may include extraneous data in the form of its background. The background may include the thermal output of machinery and equipment present in the area as well as other carcasses or even human workers that may be present.

It can be readily seen that background equipment, in the form of metalwork and piping (300) is visible in the thermal images of Figures 3a and 3b.

Therefore, a threshold temperature may be defined to filter the background and allow better definition of the carcass portion of the image. For example, this may be set at 25°C; any portion of the image at a lower temperature is appropriately filtered. This filtering may involve the segregation of the area of the 16 image (224) captured by the PIR camera (204) into blocks. A filter may be applied to further refine the 2016102045 28 Nov 2016 image to identify the cut edge of the pelt (114).

In particular, cluster analysis (using k-means) is performed. This divides the data into two clusters; one for flesh (116) and the other for the pelt (114). It has been found that two-clusters provide good results but three-clusters are superior and is preferred if it is not possible to adequately exclude extraneous background data.

Figures 4a and 4b shows the thermal images of Figures 3a and 3b following the application of a three-cluster analysis. In this image the temperature difference between the pelt and the flesh has clearly been defined. From this, the demarcation between the pelt and the flesh is readily visible.

However, this is not meant to be limiting and other algorithms suitable for use in differentiating the pelt of a carcass from the underlying flesh would be readily apparent to a person skilled in the art.

The image may be overlaid with data obtained from the three-dimensional imagery system (206) shown in Figure 2. This allows the flesh (116) and pelt (114) to be mapped with co-ordinates given for specific locations and in particular, gripping locations.

For example, the height of the brisket may be determined and from this the gripping points on the forelegs and neck are defined, as shown in Figure 5. Returning to Figure 2 showing the schematic of the system (200), a signal (228) containing the co-ordinates for the gripping locations can then be sent to the robotic arms (212). This in turn will direct the clamping mechanism (210) to the appropriate gripping locations and articulate them such that the pelt (114) is firmly held. Having achieved a secure grip, the arm is then articulated in the appropriate manner to remove the pelt from the carcass (102).

The entire disclosures of all applications, patents and publications cited above and below, if any, are herein incorporated by reference. 17

Reference to any prior art in this specification is not, and should not be taken as, an acknowledgement 2016102045 28 Nov 2016 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.

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.

Aspects of the present invention have been described by way of example only and it should be appreciated that modifications and additions may be made thereto without departing from the scope thereof as defined in the appended claims. 18

Claims (7)

1. WHAT WE CLAIM IS:

   1. A method of detecting and locating a fresh cut on an animal carcass, the method including the steps of: a) defining a processing area which encompasses at least part of a surface of the carcass which has been mechanically worked by a cutting implement; b) detecting the thermal output from that surface; c) analysing differences in thermal output from that surface to identify a fresh cut on the surface; d) taking a three-dimensional image of the processing area; e) mapping the coordinates of the identified fresh cut; and f) sending the coordinates of the identified fresh cut to an industrial robot with one or more robotic arms provided with a clamping or gripping mechanism to remove at least a portion of a pelt from the carcass.
2. 2. The method as claimed in claim 1, wherein the co-ordinates are for an exposed edge of the pelt covering a portion of the carcass and which are potential gripping points for the clamping or gripping mechanism.
3. 3. The method as claimed in claim 1, wherein the thermal output from the surface of the carcass that has been mechanically worked is detected by taking a thermographic image using a thermographic camera, wherein the thermographic camera detects passive infrared radiation.
4. 4. The method as claimed in claim 3, wherein the three-dimensional image is taken with a stereo camera which detects structured infrared.
5. 5. The method as claimed in claim 4, wherein the thermographic image is mapped to the threedimensional image to obtain the co-ordinates of the identified fresh cut.
6. 6. A system for detecting and locating a fresh cut on an animal carcass, the system including: a camera configured to detect thermal output; a processor linked or otherwise connected to the camera; three-dimensional imaging system communicative with the processor; an industrial robot with one or more robotic arms to which tooling is mounted, and wherein the processor is configured to execute the following steps: a) defining a processing area which encompasses at least part of a surface of the carcass which has been mechanically worked by a cutting implement; b) detecting the thermal output from that surface with the camera; c) analysing differences in thermal output from that surface to identify a fresh cut on the surface; d) taking a three-dimensional image of the processing area; e) mapping the coordinates of the identified fresh cut; and f) sending the coordinates of the identified fresh cut to the industrial robot with one or more robotic arms configured with a clamping or gripping mechanism to remove a pelt from the animal carcass.
7. 7. The system as claimed in claim 6, wherein the camera detects passive infrared radiation and wherein the system includes a stereo camera to take the three-dimensional image of the processing area.