Method for tracking food in a processing facility and system for processing food

JP2023551471A5Inactive Publication Date: 2024-12-06MAREL SALMON
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023532155
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2020-12-04
Filing Date
2021-12-03
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing food tracking methods in processing facilities face challenges due to tags or labels separating from food products, which are aesthetically undesirable and unreliable, while printed marks may not be suitable for all applications, necessitating a more robust and non-invasive identification method.

Method used

A method and system for automatically tracking food products using unique identification symbols (UIIs) generated from check-in and checkout images, allowing image-based identification through conveyor systems, cameras, and processing stations, with algorithms to match and associate images across different stages of processing.

Benefits of technology

Enables reliable and efficient tracking of food products without physical tags, ensuring accurate identification and metadata association, facilitating categorization and quality control throughout the processing facility.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A method and system for tracking food products in their flow through a processing facility. The food products are received at a check-in location where a check-in image is taken and a unique identification, UII, is generated based on the check-in image. The food products are processed and moved to a check-out location where a check-out image is taken. To facilitate automated tracking of the food products, a recognition process is performed to establish whether the check-out image can be assigned to the generated UII, and in that case a data record is generated with the recognized UII.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to methods and systems for tracking food passing through a processing facility, particularly to methods in which food is received in a food stream and a check-in image is acquired at a check-in location.

Background Art

[0002] In a food processing factory, food is often sequentially conveyed to workstations for trimming, filleting, sorting, or bone removal, etc.

[0003] Workstations are often arranged along a conveyor, and processing means such as operators, robots, or knives can access the food either while the food is moving or while the food is placed on an intermediate processing table after processing and returned to the conveyor.

[0004] It is generally known to uniquely identify food during processing. In one example, animals raised for meat are usually tagged with a barcode, number, or other form of unique identification tag that can be electronically read. Such identification enables the identification of data related to weight, fat percentage, quality, and processing lines within the processing facility. If necessary, such data can be useful, for example, to track contamination or to generally meet customer requirements regarding food certification.

[0005] Tags, labels, and similar identifiers applied separately can be inadvertently separated from the food during processing and thereby become useless, while printed marks, embossed stamps, and similar physical markings applied directly to the food may be undesirable, for example, for aesthetic reasons.

Summary of the Invention

[0006] To improve the ability to track food during processing, the Disclosure provides, in a first embodiment, a method for automatically tracking food in the flow of food through a processing facility.

[0007] This method, At the check-in location, receive the food item. Obtain at least one check-in image of the aforementioned food item, A unique identifier (UII) is generated from the aforementioned check-in image. In the flow of food through the processing facility, the food or a piece of food is moved to the checkout position. At the aforementioned checkout location, obtain at least one checkout image of the food or a piece of the food. A recognition process is executed which is configured to determine whether the checkout image can be assigned to the generated UII. If it is established that the generated UII can be assigned to the checkout image, define a data record that includes the assigned UII. To prepare for it.

[0008] In particular, food can be moved within the processing facility by a conveyor system that forms a path from a camera providing a check-in location and a check-in image of the food, through a processing station, to another camera providing a check-out location and a check-in image.

[0009] The recognition process involves determining, from at least one checkout image, whether that checkout image can be assigned to one of the generated UIIs.

[0010] Because the UII is recognized in the checkout image, the image of the food itself serves as the identification, eliminating the need for tags, labels, or similar physical identification marks between the check-in and checkout locations.

[0011] The data record may be stored in a database or affixed to the food or food packaging at the checkout location.

[0012] Foods may be items from a list consisting of, for example, vegetables, fruits, meat, poultry, fish, and seafood. Foods may be cut into multiple pieces. That is, in this specification, the term "piece" refers to a portion cut from food. The term "portion" refers to an area of ​​food that has not necessarily been cut from food and therefore does not necessarily need to form a piece of food.

[0013] The UII may be a numerical format, such as a data string containing integers, and may be defined by a process that uses an algorithm to convert an image to such a numerical format.

[0014] Check-in and / or check-out images may be captured using imaging means known in the art, including any type of camera that captures electromagnetic radiation, such as a line-type or matrix-type CCD camera, or a camera that captures electromagnetic radiation outside the visible spectrum, such as an X-ray camera. Check-in and / or check-out images may also be captured using ultrasound.

[0015] The processing station may include one or more stations or combinations of stations for sorting, weighing, cutting, slicing, trimming, pinbone removal, batching, packaging, or simply counting the food, or for registering the food. Thus, the food may be cut into multiple pieces after passing through the processing station.

[0016] Food can be moved, in particular, using a conveyor system, such as a conveyor belt.

[0017] In this specification, "recognize the generated UII from at least one checkout image" means that at the checkout location, the food is assigned to the UII originally assigned based on the checkout image.

[0018] In one embodiment, an additional UII, referred to herein as the checkout UII, is generated based on the checkout image. The checkout UII is generated using the same method, for example, the same algorithm and procedure used to generate the UII based on the check-in image. The checkout UII is then compared to the UII.

[0019] The recognition process recognizes that the checkout image can be assigned to one of the generated UIIs if the checkout UII matches, or at least partially matches, the UII.

[0020] If the UII and the checked-out UII are, for example, numerical values ​​contained in a data string, the recognition process may include comparing the numerical values ​​or data strings and evaluating a specific amount of distance or overlap.

[0021] In one example, a check-in image is converted into a first array of numbers, and this process is repeated to determine a second array of numbers using a check-out image. As a result, the second array may have, for example, all or some of the digits identical, such as 12 out of 15 digits. The recognition process may include defining a threshold for the number of overlapping digits; for example, a threshold of at least 10 out of 15 digits is the limit for which a UII is considered "recognized." If fewer than 10 out of 15 digits are identical, the recognition process may return "not recognized," otherwise it may return a recognized UII.

[0022] The method may include the step of combining metadata related to a processing facility or metadata related to a food, such as data received by the food, i.e., data obtained before entering the check-in location, with the UII.

[0023] Examples of such metadata may include data derived from the farm or the location where the food was originally produced, or simply identifiers of such farms or locations. The metadata may include information related to the weight, size, shape, or quality of the food, and may include, for example, the use of pesticides on vegetables or vaccines used on animals.

[0024] Examples of metadata related to a processing station may include identification of the equipment used in processing, identification of the operator processing or handling the food, and may specify environmental parameters such as temperature or humidity.

[0025] In one example, the metadata is capable of the performance of the processing station, such as dividing incoming food or pieces of food into different categories (e.g., some food or pieces of food are graded differently, some food or pieces of food are rejected) exiting the facility at different checkout locations.

[0026] The metadata and the UII may also be combined in a data file. Thereby, an image of the food or a piece of food is associated with the metadata.

[0027] The method may further include the step of defining a further processing station and the step of selecting a defined processing station for the food or a piece of food and generating a data file including the identification of the selected processing station and the UII. The selection of the further processing station may be based, for example, on the information in the metadata.

[0028] As used herein, the terms "processing station" and "processing station" may be used interchangeably unless otherwise defined.

[0029] For example, the metafile may indicate that a food product with a specific UII generated based on the check-in image has certain quality parameters that justify packaging without further processing such as trimming, or it may specify that the food product is unacceptable, etc.

[0030] The method may further comprise selecting a specific one of a set of further processing stations and assigning that specific one to a food item or food item when the UII cannot recognize it from at least one checkout image. A food item may not be recognized, for example, at the checkout location, in which case food identification is unavailable. For example, it may be determined whether such food is sold at a low price or used in a different context, such as a pre-processed food, for example, in a cooked state.

[0031] This method may include the step of identifying undesirable food portions in a check-in image and generating a UII from the portion of the image that does not contain the undesirable food portions. Examples of such undesirable food portions include portions containing a lot of fat, bone, cartilage, etc., portions with undesirable coloring, such as bloodstains, foreign objects such as blue plastic or needles, or food defects.

[0032] To enable UII recognition based on checkout images, it is sometimes desirable to generate the UII based on more image information than is necessary to determine the UII, thereby making the recognition process robust to changes between check-in and checkout images. Such processes exist in the art and generally rely on redundant image data included for UII generation and subsequent UII identification. In one example, the UII consists of at least 15 digits, but may be generated with 20 or more digits to make recognition more robust.

[0033] This method may include establishing multiple UIIs for each food item. UIIs may be established, for example, for different parts of the food item. This allows for the identification of the UII of individual fragments, for example, if an undesirable portion is cut off, or if the food item is cut into fragments at a processing station between the check-in and check-out locations.

[0034] This method may, for example, include defining a portioning plan in the check-in image that defines how food is cut into pieces, and generating a UII for each piece of food. Each piece of food can then be individually recognized at the check-out location.

[0035] The method may include generating a UII based on the structural uniqueness of the food. Examples of structural uniqueness of food include patterns of fat, bone, or muscle fibers, cartilage, muscle, and other food-specific structures. The UII may be generated based on the size or shape of the food, based on the color or color distribution of the food, or a combination thereof. In one example, the UII is generated based on the thickness profile of the food, which is determined by a check-in image taken by a line scan camera while the food is being transported on a conveyor belt.

[0036] This method may include the step of establishing a queue of UIIs for food items whose check-in images are obtained at the check-in location, and removing UIIs recognized in the check-out image from the queue.

[0037] If multiple UIIs are generated for a single identical check-in image, these UIIs may be grouped into a set of UIIs, and the method may include the step of removing all UIIs from the queue of a set of UIIs when a process is established to recognize that one or more check-in images can be assigned to a predetermined number of UIIs in that set of UIIs. The predetermined number may be "all UIIs in the set of UIIs" or a certain percentage, for example, when 90% of the UIIs in the set of UIIs are recognized in the check-out image. This allows the queue to be reduced once the majority of the fragments derived from the food have passed the check-out location.

[0038] By assigning a check-in time to each UII, it is possible to establish when a UII was generated. Subsequently, the UII may be removed from the queue after a predetermined time has elapsed. This reduces the number of UIIs in the queue to a level that includes only those UIIs to which the queue is relevant, thereby enabling a faster and more efficient recognition process.

[0039] The queue defines a movement window in the checkout image where the amount of potentially recognizable UIIs is relatively small, limiting the risk of recognizing incorrect UIIs. In particular, if the recognition process includes an evaluation of the best match between the UII and the checkout UII, it can improve the ability to provide correct recognition even when the UII and the checkout UII do not necessarily match.

[0040] This method may be applied to a wide variety of foods, such as vegetables, fruits, meat, poultry, fish, and seafood. In particular, this method may be used when processing meat such as fish fillets, such as salmon fillets, meat, beef, pork, poultry fillets, or slices thereof.

[0041] In one embodiment, food or food fragments are classified into at least a first category and a second category. The classification is performed between the check-in and check-out locations and may be associated with different quality standards, for example. In one embodiment, food or food fragments of the first category are led to the check-out location, while food or food fragments of the second category are not led to the check-out location.

[0042] For example, food or food fragments in Category 2 may be excluded and transported to other checkout facilities. The facility may have, for example, two, three or more different exits related to different categories.

[0043] A queue of UIIs for food items whose check-in images have been acquired at the check-in location may be established. This queue may be the aforementioned queue, i.e., a queue that is updated when a food item or food item passes through the check-out location or when the timeout period expires.

[0044] Food or food fragments may be removed from the queue if they are classified as belonging to a category other than the second category or another category that is not intended to pass through the checkout location. In this case, the food may pass through an alternative checkout location. This feature may be used, for example, when food or food fragments fail to pass between the check-in and checkout locations. Here, “rejected” (or “excluded”) means “not suitable for a particular use,” and therefore may include downgrading the food or food fragments for another use, which may include edible use.

[0045] In addition to generating a data record with the assigned UII for food or food fragments to which a checkout image can be assigned, the method may also include defining a data record for all food items passing through the checkout location. In this case, a checkout ID is defined for each and all food items at the checkout location, and the data record may include the checkout ID and optionally include the UII if a checkout image can be assigned to the generated UII. If no UII is assigned, the data record may simply be omitted or may include information that no UII was assigned.

[0046] In a second aspect, the disclosure provides a system for processing food having an automated tracking capability. Such a system A conveyor configured to move the food from a check-in location to at least one processing station and a check-out location, A check-in camera is positioned at the aforementioned check-in location and configured to capture a check-in image of the food item. A processing structure configured to generate unique identifiers (UIIs) from the aforementioned check-in image, A checkout camera is positioned at the checkout location and configured to capture checkout images of the food or a piece of the food. It is equipped with.

[0047] The processing structure is configured to execute a recognition process that determines whether a checkout image can be assigned to the generated UII, and if it is determined that a checkout image can be assigned to the generated UII, it is configured to define a data record containing the assigned UII, for example, in combination with the checkout ID.

[0048] Corresponding to the method of the first embodiment, the processing structure may define a data record for each and all food or food fragment that passes through the checkout location, and the UII is included in or omitted from the data record depending on whether the UII is recognized or not.

[0049] The processing structure may be configured to establish a queue of UIIs for food items whose check-in images were obtained at the check-in location, and to remove UIIs recognized in the check-out image from the queue.

[0050] The system may be further configured to assign a check-in time to each UII when it is generated, and to remove the UII from the queue after a predetermined time has elapsed.

[0051] The system is configured to determine the lead time for food items from the check-in point to the check-out point, and this predetermined time is dynamically updated based on the lead time.

[0052] The system may include a conveyor structure that allows for the formation of a gap between a feeding section and a discharging section, where the feeding and discharging are relative positions to the gap between the two sections, and the feeding and / or discharging sections are movable relative to the gap. In relation to the gap, a knife may be positioned so that during a cutting process for cutting / dividing food, the cut pieces may be guided into the gap, or the end portions of the food may be guided into the gap.

[0053] Such a structure may allow for the removal of food or food fragments at a location between the check-in and check-out points, regardless of whether the process is performed in between. In the case of salmon processing and general fish processing, food may be the whole fish, and food fragments may include the tail and head, which are typically removed between the check-in and check-out points.

[0054] Furthermore, the system relating to the second aspect may include any features implicit to the method relating to the first aspect of this disclosure.

[0055] (List of embodiments) [Embodiment 1] A method for tracking food in the flow of food through a processing facility, Upon receiving the aforementioned food, Move the aforementioned food item to the check-in location. Obtain at least one check-in image of the aforementioned food item, At least one unique identifier (UII) is generated from the aforementioned check-in image, The food is moved from the check-in location to at least one processing station. Move the food or a piece of the food from the processing station to the checkout location. At the aforementioned checkout location, obtain at least one checkout image of the food or a piece of the food. Recognizing the UII generated from the at least one checkout image A method for preparing for something.

[0056] [Embodiment 2] The method according to Embodiment 1, comprising generating a plurality of UIIs from the aforementioned check-in image.

[0057] [Embodiment 3] The method according to Embodiment 2, wherein the plurality of UIIs include UIIs related to a portion of the check-in image and other UIIs related to a different portion of the check-in image, i.e., a different portion of the food.

[0058] [Embodiment 4] The aforementioned differing parts are overlapping parts, according to the method of Embodiment 3.

[0059] [Embodiment 5] The method according to any one of embodiments 2 to 4, further comprising defining a set of UIIs for each food item that define the relationships between UIIs belonging to the same food item.

[0060] [Embodiment 6] The method according to any one of embodiments 1 to 5, comprising establishing a queue of UIIs of food items for which a check-in image was obtained at the check-in location, and removing UIIs recognized in the check-out image from the queue.

[0061] [Embodiment 7] The method according to embodiment 5 or 6, further comprising deleting all UIIs from the set of UIIs when a predetermined number of UIIs are recognized in the checkout image.

[0062] [Embodiment 8] The check-in time established when the aforementioned UII is generated is assigned to each UII. The method according to embodiment 6 or 7, further comprising removing the UII from the queue after a predetermined time has elapsed.

[0063] [Embodiment 9] The method according to any one of embodiments 1 to 8, wherein the UII is combined with metadata associated with the processing station, metadata received by the food, or metadata obtained at the check-in location.

[0064] [Embodiment 10] The method according to any one of embodiments 1 to 9, comprising determining a further processing station for processing the food after the checkout location, selecting a designated processing station for the food or a fragment of the food, and generating a data file including the recognition of the selected processing station and the UII.

[0065] [Embodiment 11] The selection is according to the method of Embodiment 9 or 10, based on the information in the metadata.

[0066] [Embodiment 12] The method according to Embodiment 10 or 11, comprising selecting a specific one of the designated further processing stations and assigning the specific one to the food or a piece of the food when the UII cannot be recognized from at least one checkout image.

[0067] [Embodiment 13] The method according to any one of embodiments 1 to 12, comprising identifying undesirable food portions in the check-in image and generating at least one UII from the portion of the image that does not contain the undesirable food portions.

[0068] [Embodiment 14] The method according to any one of embodiments 1 to 13, comprising identifying undesirable food portions in the check-in image and generating at least one UII from the portion of the image containing the undesirable food portions.

[0069] [Embodiment 15] The method according to Embodiment 14, comprising identifying at least one UII from the portion of the image containing the portion of the undesirable food, and eliminating the food or the portion of the food containing the undesirable food based on the recognition.

[0070] [Embodiment 16] The method according to any one of embodiments 1 to 15, comprising determining a portioning plan that defines how the food will be cut into pieces based on the check-in image, and generating at least one UII for each of the pieces of food.

[0071] [Embodiment 17] The method according to any one of Embodiments 1 to 16, wherein the UII is generated based on the structural characteristics of the food, based on the size or shape of the food or a piece of the food, based on the color or color distribution of the food, or a combination thereof.

[0072] [Embodiment 18] The method according to any one of Embodiments 1 to 17, wherein the food is selected from the group consisting of vegetables, fruits, meat, poultry, fish, and seafood, or slices thereof.

[0073] [Embodiment 19] The method according to any one of Embodiments 1 to 18, wherein the food is salmon such as salmon fillets.

[0074] [Embodiment 20] The method according to any one of embodiments 1 to 19, wherein the food is classified into at least a first category and a second category, the classification is performed between the check-in position and the check-out position, the food or food fragments of the first category are led to the check-out position, and the food or food fragments of the second category are not led to the check-out position.

[0075] [Embodiment 21] The method according to Embodiment 20, comprising establishing a queue of UIIs for food items whose check-in images have been acquired at the check-in location, and removing UIIs for food items or food fragments classified into the second category from the queue.

[0076] [Embodiment 22] A food processing system having automatic tracking capabilities, A conveyor configured to move the food from a check-in location to at least one processing station and a check-out location, A check-in camera is positioned at the aforementioned check-in location and configured to capture a check-in image of the food item. A processing structure configured to generate unique identifiers (UIIs) from the aforementioned check-in image, A checkout camera is positioned at the checkout location and configured to capture checkout images of the food or a piece of the food. Equipped with, The processing structure is configured to recognize the generated UII from at least one of the checkout images.

[0077] [Embodiment 23] The system according to Embodiment 22, wherein the processing structure is configured to establish a queue of UIIs of food items for which check-in images have been acquired at the check-in location, and to remove UIIs recognized in the check-out image from the queue.

[0078] [Embodiment 24] The system according to embodiment 23, wherein the processing structure is configured to assign a check-in time to each UII when each UII is generated, and to remove the UII from the queue when a predetermined time has elapsed.

[0079] [Embodiment 25] The system according to Embodiment 24, comprising a sensor configured to determine the lead time of the food from the check-in location to the check-out location, and the predetermined time before the check-out is dynamically updated based on the lead time.

[0080] (List of drawings) The embodiments of this disclosure will be described in more detail below with reference to the drawings. [Brief explanation of the drawing]

[0081] [Figure 1] Figure 1 schematically shows the main components of the system relating to this disclosure. [Figure 2] Figure 2 shows further details of the system related to this disclosure. [Figure 3] Figure 3 shows foods with identifiable structural characteristics. [Figure 4] Figure 4 shows areas of low quality, such as food containing undesirable elements. [Figure 5] Figure 5 shows the food items from Figure 4 with the portioning plan superimposed. [Figure 6] Figure 6 shows food items that generated multiple UIIs in the checkout image. [Figure 7] Figure 7 shows the timeline. [Modes for carrying out the invention]

[0082] The detailed description and specific examples illustrate embodiments, but are for illustrative purposes only, as the spirit and scope of this disclosure will become apparent to those skilled in the art from the detailed description.

[0083] Figure 1 schematically shows a system comprising three stations named S1, S2, and S3. The three stations may be connected, for example, by a conveyor belt, or they may simply be three stations located at three different positions on a table. S1 may represent the check-in position. S2 may represent a processing station, for example, a processing station that processes food or processes it in other ways. Examples of such processing include sorting, counting, marking, filleting, trimming, weighing, batching, or any similar known processing of food. S3 may represent the check-out position.

[0084] Figure 2 shows an example of such a system schematically illustrated in Figure 1. In Figure 2, system 201 is shown in more detail. The illustrated system is configured for operator-assisted processing of food 202, but may be configured for fully automated processing or for simple processing of food (e.g., simple registration of food for packaging).

[0085] The illustrated system has an automatic tracking function that allows each food item that enters the system to be recognized when it leaves the food processing system. The illustrated facility includes a conveyor 203 that transports food from an inlet 204 to an outlet 205.

[0086] The check-in position 206 is defined along the conveyor. At this position, a check-in camera 207 is positioned to capture a check-in image of the food. In the illustrated embodiment, the check-in camera is positioned above the conveyor, and a lamp 208 may be positioned to provide sufficient light to recognize the details of the food. The camera 207 is based on visible light reflection, but in alternative embodiments, the camera may be X-ray based or ultrasonic based, etc.

[0087] A checkout camera 209 is positioned at the checkout location 210 and, with the assistance of, for example, a lamp 211, can capture checkout images of food or food fragments. Cameras 207 and 209 may be positioned in front of the corresponding lamps 208 and 211, but the lamps may be positioned in front of the cameras or at any other suitable location.

[0088] Between the check-in and check-out locations, food is processed at a processing station. In this case, the processing station comprises two processing tables 212 and 213. Alternatively, such a processing station may be equipped with an automated processing device.

[0089] Operators 214 and 215 may be assigned to each processing table. Each table and each operator has an individual ID, for example, in the form of an identification number.

[0090] Computer 216 includes a data input unit configured to receive images from check-in and check-out cameras, for example, via a local area network, i.e., LAN 217. A CPU having corresponding software code is configured to form a processing structure, generate a unique identifier (UII) based on the check-in image, and identify the UII in the check-out image. Computer 216 may generate a data record 218, symbolized by a product card 218. The data record may include the UII and other types of data, such as metadata related to food or food processing, such as an operator ID.

[0091] Each table may have a computer interface, for example in the form of a touchscreen 219,220, which allows an operator to identify food and generate information related to the food. Such information may identify the operator's ID, the table's ID, or visually observed problems related to the food, such as quality parameters observed by the operator. The information may be transmitted via LAN 217 to a computer 216 or other location, such as a supervisor computer system 221 where a supervisor 222 can review the data. The data may be included as metadata in a data record 218. The data, such as that included in a data record, may be exported, for example, to an adjacent processing station, or to track food or food fragments all the way to the consumer.

[0092] An outlet may be located to deliver food or food fragments to an adjacent processing station for, for example, further processing, packaging, inspection, or disposal.

[0093] The system may include a diverter that can be controlled by a processing structure so that, depending on the recognition of food or food fragments at the checkout location, the food is directed to one or the other of subsequent processing stations. For example, the processing structure may have a list of N specific food items, each having a corresponding UII. If these N food items are to be packaged for a specific customer, the food diverter may be controlled by the processing structure so that these specific food items are directed to a packaging station dedicated to that customer. If the specific food item cannot be recognized at the checkout location, the processing structure may use a pre-selected subsequent processing station, for example, a processing station with a human operator capable of determining the action for the unrecognized food item.

[0094] UII may be generated based on the structural characteristics of the food. Examples of such structural characteristics include the shape and size of fat, bone, muscle, and fiber patterns. UII may also be generated based on the shape or size of the food, or based on the color or color distribution of the food.

[0095] In particular, the check-in image may be a food image prepared to obtain the best possible image of this unique feature.

[0096] Figure 3 illustrates a fillet of fish, specifically salmon. The illustrated salmon has been filleted. This process reveals the structure of the meat and fat, which can be used to obtain a good image representing a unique "fingerprint" that defines the UII. If filleting is desired, the system may include a filleting station prior to the check-in position, in particular to make a check-in image, including the fingerprint structure of the fillet, available.

[0097] The gray area indicated in box 301 is typically cropped and removed for further processing. The salmon fragments within the gray area 301 may be classified as food that is not intended to pass through the checkout location. Therefore, the aim may be to remove UIIs from such food fragments from the UII queue to facilitate faster or more robust identification of the remaining UIIs in the checkout image.

[0098] The food shown in Figure 4 is photographed in a check-in image. The processed structure identifies bloodstain 402. Such bloodstains are generally undesirable and are therefore usually cut off during processing.

[0099] Since the above cropping may be performed between the check-in and check-out points, it is desirable to either not use the area around the bloodstain for the purpose of generating UII, or to generate at least UII based on a principle that is robust to the removal of parts of the food, i.e., parts of the image.

[0100] If such undesirable elements of food are removed, the processing structure's ability to recognize UIIs in the checkout image may be impaired. Therefore, when the processing structure identifies such undesirable parts of food, it is programmed to generate a frame 403 around the bloodstain 402 and generate UIIs from the portion of the image that does not contain the frame 403.

[0101] Figure 5 illustrates the food from Figure 4, with a portioning plan superimposed, defining the cutting lines 501 for cutting the food into portions. The portioning plan is used to control the portioning cutter positioned between the check-in and check-out positions. Due to this process, it can be difficult to recognize the UII, and therefore the processing structure is configured to generate the UII for each portion according to the portioning plan.

[0102] Figure 6 illustrates a food item in which multiple UIIs were generated during the check-in image. The generated UIIs are uniquely numbered 61-90 and relate to different unique parts of the food item. In an alternative embodiment, at least some of the UIIs may relate to overlapping parts of the food item.

[0103] Computer 216 may define a set of UIIs that define the relationships between UIIs belonging to the same food for all of these UIIs related to the same food. For example, all UIIs 61-90 may be defined as UIIs 1-61, 1-62, 1-63...1-90, thereby defining that they belong to food number 1.

[0104] Figure 7 shows a timeline T illustrating the lead time, as defined herein, from the time a food item enters the check-in location S1 until it exits the check-out location S3. Depending on the specific layout of the system, a finite number of food items may be processed, for example, simultaneously or sequentially. To improve the efficiency of the system, and in particular to enhance the processing structure's ability to recognize UIIs generated from check-out images, it may be beneficial to reduce the number of food items considered by the processing structure.

[0105] In Figure 7, five food items labeled A enter check-in position S1. Seven food items labeled B are being processed. These five are in the flow, and two of the seven food items, labeled B', are currently waiting between check-in position S1 and check-out position S3. The reason for waiting for food items is, for example, that the processing needs for incoming food items are uneven, resulting in insufficient available resources. If multiple consecutive food items require specific processing, the processing capacity may necessitate waiting for the food items until processing becomes possible.

[0106] Computer 216 has memory in which a queue of UII values ​​for food items designated with the letter B is stored.

[0107] Each time a food item arrives at the check-in location, the queue is updated with the new UII. Each time a food item leaves the check-out location and is recognized from the check-out image for having a specific UII, the corresponding UII is removed from the queue. In this way, the queue contains only a limited number of available UIIs, which can improve the ability to accurately recognize UIIs. This feature is particularly relevant when UIIs are generated based on features, such as structure, size, shape, or color, which are considered statistically good uniqueness for relatively small quantities of food but not for large quantities.

[0108] Figure 7 shows on timeline T that the computer has a built-in clock. This clock illustrates the function of a processing structure that assigns a check-in time to each UII when each UII is generated. Based on knowledge of a particular layout, it may be determined that food has normally left the system at a predetermined time. Therefore, the check-in time may be used to remove a UII from the queue when a predetermined period has elapsed. This could happen, for example, if a waiting food B' is removed for some reason.

[0109] In one embodiment, the check-in and check-out cameras are CCD cameras positioned along a conveyor belt. The cameras communicate digital images with a processing structure. The processing structure has an image recognition unit that provides a converter capable of converting digital images from the check-in or check-out camera into unique data records. The communication unit can output the data records to another computer, for example, to recognize labels, to later authenticate food items, or to store the data records in a database.

[0110] Internally, data records are stored in memory, and a microprocessor-type image signal processing unit is configured to perform a reconstruction process to verify whether the checkout and check-in images are from the same food item. The microprocessor may be programmed with software developed using standard software or similar techniques developed for recognizing features in images. Such systems have been widely developed, for example, for face detection or fingerprint recognition.

[0111] The processing structure may include a video card having a frame buffer and other functions configured to process input from the camera, and the UII may be generated based on standard processor functions available for, for example, face recognition, fingerprint recognition, etc. Such a processor may provide a unique data string representing the UII of the image.

Claims

1. 1. A method for tracking food products in their flow through a processing facility, comprising: receiving said food items at a check-in location; acquiring at least one check-in image of the food item; generating a unique identification code (UII) from the check-in image; moving the food products or food product pieces to a checkout location in a flow of the food products through the processing facility; acquiring at least one checkout image of the food item or food piece at the checkout location; executing a recognition process configured to establish whether the checkout image can be assigned to the generated UII; If it is established that the checkout image can be assigned to the generated UII, a data record including the assigned UII is defined. A method comprising:

2. defining a first processing station for further processing of the food product or food product pieces; defining a second processing station for further processing of the food product or food product pieces; selecting one of the first and second processing stations when the recognition process establishes that the checkout image is assignable to the generated UII; selecting the other of the first and second processing stations if the recognition process establishes that the checkout image cannot be assigned to the generated UII. The method of claim 1 further comprising:

3. generating a plurality of UIIs from the check-in image; If the plurality of UIIs are generated from the same check-in image or if the plurality of UIIs are generated from the same food item, group the plurality of UIIs into a set of UIIs. The method of claim 1 or 2, comprising:

4. establishing one or more food queues in said stream of food; If the recognition process establishes that a checked-out image can be assigned to a UII, then remove that UII from the queue. The method according to any one of claims 1 to 3, comprising:

5. 5. The method of claim 3 or 4, comprising deleting all UII from a set of UII if the recognition process establishes that one or more checkout images are assignable to a predetermined number of UIIs from the set of UIIs.

6. assigning to each UII a check-in time that is established when said UII is created; Remove the UII from the queue when a predetermined time has elapsed from the check-in time. The method of claim 4 further comprising:

7. Identifying undesirable food portions in the check-in image; generating at least one UII from a portion of the image that does not include the undesirable food portion; The method of any one of claims 1 to 6, comprising:

8. Identifying undesirable food portions in the check-in image; generating at least one UII from a portion of the image that includes the undesirable food portion; The method of claim 1 , comprising:

9. 9. The method of claim 1, further comprising: assigning a grade to the food product or food piece depending on whether the recognition process establishes that a checkout image is assignable to a UII or whether the recognition process does not establish that a checkout image is assignable to a UII.

10. defining a portioning plan based on the check-in image, the portioning plan defining the intended cutting of the food product into portions of the food product; generating at least one UII for each of the pieces of the intended food product; The method of any one of claims 1 to 9, comprising:

11. each food item or each food piece in the stream of food items is classified into at least a first category and a second category, said classification being performed between the check-in location and the check-out location; 11. The method of claim 1, wherein food items or food pieces of the first category are directed to the checkout location and food items or food pieces of the second category are not directed to the checkout location.

12. The method of claim 11 , wherein the second category of food items or food pieces are directed to an alternate checkout location.

13. establishing a UII queue for the food item for which a check-in image was captured at the check-in location; Remove UII for food or food pieces classified in the second category from the queue. The method according to claim 11 or 12, comprising:

14. defining a data record for each food item or each food piece at the checkout location; if it is established that the checkout image can be assigned to the generated UII, assigning the UII to the food or a portion of the food; 14. The method according to any one of claims 1 to 13.

15. 1. A system for tracking food products in their flow through a processing facility, comprising: a check-in camera disposed at a check-in location of the processing facility and configured to capture a check-in image of the food product; a processing architecture configured to generate unique identifiers (UII's) from the check-in images; a checkout camera disposed at a checkout location of the processing facility and configured to capture a checkout image of the food product or food piece; Equipped with The processing structure is configured to perform a recognition process to establish whether the checkout image can be assigned to the generated UII, and if it is established that the checkout image can be assigned to the generated UII, the system is configured to define a data record including the assigned UII.