Method for introducing food blocks into a food processing and processing line or device for carrying out such a method
The method uses optical detection and data evaluation to inspect packaging residues on food blocks, ensuring reliable detection and removal, thus preventing contamination and enhancing processing efficiency.
Patent Information
- Application Number
- DE102024110736
- Authority / Receiving Office
- DE · DE
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-04-17
- Publication Date
- 2025-10-23
AI Technical Summary
Existing methods for checking food blocks for packaging residues are prone to errors and can lead to contamination in the processing of finished products, necessitating costly recalls.
A method involving optical detection and electronic data evaluation to inspect the integrity of removed packaging after depackaging, comparing captured images with desired values to classify food blocks as non-critical or critical, with additional manual or automated checks for residues.
Ensures reliable detection and removal of packaging residues, reducing the risk of contamination and enabling efficient processing of food blocks.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a method for introducing food blocks into a food processing system according to the preamble of claim 1. The invention further relates to a processing plant or a device for carrying out such a method.
[0002] The generic process is characterized in particular by the fact that packaged, deep-frozen food blocks are delivered to a processing line. These food blocks are usually wrapped in waxed paper or cardboard, the wax coating being intended to prevent the packaging material used as incoming goods packaging from adhering to the food block, for example, by freezing it to the surface. The incoming goods packaging typically serves as transport packaging and also to protect and preserve the product's shelf life.
[0003] Typical food blocks produced using the generic method are, for example, cubic or cuboid in shape and weigh between 5 kg and 10 kg, preferably 7.5 kg. They are deep-frozen to a temperature below -15°C, preferably below -20°C, and are sometimes processed while still frozen after unpacking. When this description refers to a food block in the claims, it also includes an assembly of several individual blocks into a cluster grouped together in a common incoming goods packaging.
[0004] Upon delivery, the food block is initially unpacked manually or automatically in the established processes. It is then often checked for remaining packaging material by the personnel operating the processing line and performing incoming goods inspection. This manual check, if carried out at all, is usually performed on a random sample basis and is subject to a certain degree of error even when all incoming goods are thoroughly inspected.
[0005] However, remnants of the packaging cause significant problems during the subsequent processing of the food block and, in the worst case, can even lead to the recall of an entire batch of finished products if it cannot be ruled out that individual products contain remnants of the packaging.
[0006] It is known from the prior art to automatically inspect unpacked incoming goods for quality assurance purposes. For example, WO 2018 / 236 844 A1 describes a method in which the incoming goods are observed and automatically visually inspected. A similar method is known from EP 3 801 934 B1.
[0007] Optical detection devices for optical image processing are described in EP 3 895 541 B1. However, these do not include any quality assurance measures.
[0008] The object of the invention is to subject incoming goods to the simplest possible yet reliable quality control, so that the risk of packaging residues entering the processing along with the food block is eliminated as far as possible, or at least reduced. A further object of the invention is to provide a processing plant or device with which the method according to the invention can be carried out.
[0009] The problem with regard to the method is solved according to the invention by a method according to claim 1. With regard to the processing plant or the device, the problem is solved by a processing plant or device according to claim 15.
[0010] The new process is characterized in particular by the fact that, after the removal of the incoming goods packaging, a quality control check is performed using scanning equipment and electronic data analysis. This check includes the optical scanning of the incoming goods packaging after its removal from the frozen food block and a comparison of the optical scanning result with at least one target image and / or target pixel value. For the sake of readability, this description refers to "images" in the claims. This includes all digital equivalents of a visually recognizable image of the incoming goods packaging; the only essential requirement is that the stored and captured data are suitable for a target-actual comparison. Thus, "image" also refers to pixel data, which is compared with the target data either as a whole or line by line.
[0011] According to the invention, it is not the unpacked incoming goods, but rather the empty incoming goods packaging that is checked for integrity and completeness after it has been removed from the food block. Tears or other damage that obviously do not result in any remnants of the incoming goods packaging remaining on the food can usually be disregarded. However, if entire pieces of the incoming goods packaging are missing, it is reasonable to suspect that these remain attached to the food block as packaging residue, and there is a risk that this packaging residue will enter the processing.
[0012] Furthermore, it is possible that surface areas, for example parts of the waxed inner layer, have detached as surface residue and that the packaging may appear generally undamaged from the outside, but parts of this inner layer are adhering to the frozen food block.
[0013] The invention now evaluates, in a preferred and easily performed step, the image of the incoming goods packaging, which is captured by scanning devices after the incoming goods packaging has been removed from the food block. In contrast to observing and inspecting the food block during its introduction into the processing line, this evaluation can be carried out relatively easily and cost-effectively as part of a continuous process. The scanning devices capture the appearance of all surfaces that have come into contact with the food block.
[0014] An optional inspection of the outer packaging surfaces, which do not come into contact with the food, can also be carried out to prevent any residue from detaching from the outside and potentially entering the production process, should this be possible due to the design of the processing plant. Examining the incoming goods packaging has the additional advantage of revealing any external damage that could lead to contamination or other quality issues, thus eliminating the need for a separate search.
[0015] In a data analysis, a target image or other data suitable for a target-actual comparison, such as a pixel threshold, is stored. This data is subsequently referred to collectively as the "target image," although this does not necessarily have to be an image of the incoming goods packaging. The target image corresponds to the image or pixel equivalent of an image that should be captured if the incoming goods packaging were undamaged. The data analysis performs the target-actual comparison by comparing this target image with the result of the optical capture via the scanning equipment. If no deviations are found between the result of the optical capture and the target image, this is classified as "non-critical," and the unpacked food block is released for further processing.
[0016] If, however, deviations are detected between the result of the optical inspection and the target image, the result is classified as "critical," and the food block is preferentially removed from the continuous feed to the processing line and subjected to a more detailed inspection as part of an additional check. A critical result can mean that packaging residue is adhering to the frozen food block. However, this is not necessarily the case. Such defects can also have occurred due to the initial use of damaged incoming goods packaging during packing, or due to pieces that fell off the incoming goods packaging during unpacking but did not adhere to the food block.
[0017] To distinguish food blocks that actually show signs of remnants of the incoming packaging, such as packaging residue or surface tears, from non-critical food blocks, the rejected food block undergoes an additional, thorough inspection. Since this provides some early warning, a more extensive manual inspection can be carried out by staff. Alternatively, or additionally, a more intensive automated inspection using image capture devices can be performed. Ultimately, this depends on how well the optical detection devices can differentiate the surface of the food block from the remnants of the incoming packaging.
[0018] If remnants of incoming goods packaging are found on the food block during the additional inspection, these can also be visually detected. The data analysis can include software tools capable of overlaying these detected remnants with the results of the visual inspection. This allows for a new target-actual comparison with the target image, potentially resulting in the affected food block being classified as non-critical. If parts are still missing, the food block is either completely inspected manually or processed in another way.
[0019] In addition to simply inspecting the incoming goods packaging after unpacking the food block, further measures can be taken to optimize quality assurance. Firstly, the optical detection via scanning equipment and the quality of the resulting analysis can be improved by enhancing the detectability of the inner layer of the incoming goods packaging through appropriate measures. For example, the inner layer of the incoming goods packaging, which is usually the waxed layer, can be marked with an additional color that can be detected using suitable methods. This could be a pure contrasting color or a color with special effects. For instance, there are so-called infrared colors that can only be detected by infrared cameras after being irradiated with infrared light.
[0020] For example, if the inner layer of the incoming goods packaging consists of a plastic material or a paper layer and is printed or coated with an infrared ink, an infrared camera can easily detect holes in the expected, uniform appearance after scanning the surfaces. Conversely, there are also inks that become partially transparent or completely invisible under infrared light. These inks can also be used to perform a target-actual comparison with increased reliability.
[0021] Finally, as an alternative or additional measure to improve quality, the frozen food block can also be inspected using scanning equipment after the incoming goods packaging has been removed, and the result of the optical scan can be compared with at least one target image. Here, too, the target image is stored in the data analysis, allowing any discrepancies to be identified. The appearance of the food block, which typically exhibits a very specific appearance under UV light, for example, will change significantly when using incoming goods packaging printed or coated with UV-sensitive materials. Therefore, such film residues can be detected particularly easily and reliably using the methods described above, specifically UV-dependent printing inks or coatings, and by examining the areas with an infrared camera.
[0022] In every case, the electronic comparison of the optical image results with the target image yields a data evaluation that classifies the result as "critical" or "non-critical". If the result is deemed non-critical, the food block is sent for processing and the incoming goods packaging is recycled or disposed of.
[0023] If, however, the result is deemed critical, additional quality control is activated and the affected food block undergoes a special inspection. For this to be possible, it is essential that a clear link can be established between the food block and the original packaging from which it has since been unpacked. This link can be established in various ways. For example, in a particularly simple case, after separating the food blocks from their original packaging, the order of the food blocks on the conveyor belts can be monitored, allowing for identification based on this information.
[0024] Alternatively, the incoming goods packaging can also contain an indicator or be marked with an indicator before, during, or after unpacking, which is uniquely assigned to a food block. This assignment can be made, for example, via a corresponding indicator that can be attached to the food block or to a workpiece carrier on which the food block is located after unpacking. This method is particularly suitable if quality control is not carried out immediately after removing the incoming goods packaging, but at a later time.
[0025] The inventive method is preferably used in industrial production. In this case, a large number of food blocks are unpacked sequentially, with the unpacked food blocks and the removed incoming goods packaging then being separated. Besides manual logistics, a more complex solution could, for example, involve separating conveyor belts that allow the food blocks and incoming goods packaging to be conveyed separately. The conveyor belt conveying the incoming goods packaging then feeds the packaging to the scanning devices or passes the packaging through the scanning devices. Of course, the handling and logistical management of the food blocks and the incoming goods packaging is not essential to the present invention; all other possible solutions are also relevant.
[0026] In a particularly simple case, the scanning devices can be line scanners or flatbed scanners capable of scanning the incoming goods packaging in such a way that the relevant areas are captured. Relevant areas are, in particular, those surfaces that have been in contact with one side of the food block.
[0027] The use of the aforementioned scanners has the advantage that the incoming goods packaging can be folded flat or unfolded after unpacking. This is helpful because flat incoming goods packaging, consisting of layers of film or paper, is easier to scan. Depending on the design of the incoming goods packaging, the scanners can scan one or both sides. As an alternative to line or flatbed scanners, other optical capture devices, in particular photo or video cameras or infrared cameras, can also be used. Ultimately, the only requirement is that the capture device can record an image of the relevant areas suitable for target / actual comparison in data analysis.
[0028] If a critical result is detected, the food block can be released after an additional check, including manual inspection by staff, as described above. Alternatively, the area that triggered the critical result can be automatically cut off as a layer. The scanning equipment has detected damage on the surface of the incoming goods packaging. This damage can be clearly assigned to one side of the food block if the relative position between the food block and the incoming goods packaging is also recorded, allowing the data analysis to identify the side where parts of the incoming goods packaging might be adhering.
[0029] Once the critical side has been identified, the affected food block can be fed to a cutting station or other processing station without human intervention, so that a layer is automatically removed, in particular cut off. This removal also removes any remaining portion of the incoming goods packaging. The removed material can then be further processed, for example, after inspection by personnel, it can be sent to another processing stage.
[0030] If defined pieces of a specific target weight and desired geometry are to be produced from the food block, it is advisable to consider this requirement when cutting off the edge area affected by the critical result. For example, if the food block is to be divided into integer multiples of a single block, the cut-off edge area can be chosen to be large enough so that the remaining food block can be divided into such integer multiples without any waste. A desired geometry of the final product can also be taken into account in this way.
[0031] Further features and advantages of the invention will become apparent from the following description of preferred embodiments with reference to the drawings.
[0032] The drawings show: Fig. 1 an overview of the essential process steps of the process according to the invention and Fig. 2 a schematic representation of the process steps from the overview according to Fig. 1.
[0033] In Fig. Figure 1 shows the essential process steps of the new process in the form of a block diagram. First, a number of frozen, packaged food blocks 1 are delivered to the processing line as input material. These are unpacked on site, usually by machine. A quality control check of the input material is then preferably carried out.
[0034] If necessary for later processing, the packaging material of the incoming goods packaging 3 and the food block 2 can be marked or otherwise identified after unpacking. Markings could, for example, be transponders similar to those used in manufacturing. Any other markings that allow for the later identification of an incoming goods packaging 3 and the food block 2 it previously contained are also possible.
[0035] Subsequently, the packaging material of the incoming goods packaging 3 undergoes an automated visual inspection using a scanning device 6. This can be a flatbed scanner to which the unfolded material of the incoming goods packaging 3 is fed. The scanning device captures an image of the incoming goods packaging 3. This image is then fed to the data evaluation system 7, which, as a data processing system, includes software capable of performing a target-actual comparison with known target images. These target images are preferably stored in the data evaluation system 7. Alternatively, a comparison can also be made with the incoming goods packaging that was separated from previously introduced food blocks 2. In this way, the data evaluation system 7 can also be trained to learn, so that it is not necessary to store the target images in a database.
[0036] If the first food block 2 is already contaminated with remnants of the incoming goods packaging 3, this will be noticed when scanning the incoming goods packaging of the second and, if applicable, third food block 2. In this case, it is advisable to assign the incoming goods packaging 3 to the specific food block 2 and, before further processing the scanned food block 2, to scan at least two further food blocks 2 in order to generate a reference image as a target image in a timely manner and to be temporarily stored in the memory of the data evaluation 7.
[0037] If the evaluation of the target-actual comparison through data analysis 7 reveals that the incoming goods packaging 3 is intact, meaning that no remnants of the incoming goods packaging 3 can adhere to the food block 2, the latter is sent for further processing. This further processing can consist of dividing the food block into individual portions by punching, sawing, or other methods. The individual portions can then be processed in any desired way, for example, by breading, frying, baking, or cooking, forming in die-cutting molds, or any other known method. Further processing of the complete, undivided food block 2 is also possible.
[0038] If, however, the target-actual comparison reveals that the result of the optical inspection is "critical" because there is a (too large) difference between the target image and the result of the optical inspection, the affected food block 2 is initially considered contaminated and removed from further processing. A manual or automated inspection of the affected food block 2, which belongs to the incoming goods packaging 3 identified as critical, can then be carried out. This inspection may reveal that food block 2 is safe, in which case it is reintroduced into further processing.
[0039] If, however, it turns out that residues are adhering to the affected food block 2, it can be cleaned manually so that it can then be used for further processing. Alternatively, a layer can be removed, or food block 2 can be used for a completely different purpose.
[0040] Fig. Figure 2 shows the steps described above in a schematic view. The packaged food block 1 is delivered deep-frozen at a temperature of less than -15 °C. First, the receiving packaging 3 of the cubic food block 2 is opened, and the exposed food block 2 is removed from the receiving packaging 3.
[0041] In the illustrated embodiment, the incoming goods packaging 3 exhibits two types of damage. Firstly, there is a defect 4 caused by the complete tearing of the packaging material. Secondly, only the inner layer of the waxed coating is damaged (defect 5) by partial detachment of this layer. Both types of damage, namely defect 5 caused by detachment and defect 4 caused by tearing, are detected by optical scanning devices, schematically represented here as cameras. For this purpose, the disassembled and unfolded cardboard box of the incoming goods packaging 3 is completely scanned, and the resulting image is compared with the target image stored in a data analysis system 7.
[0042] The comparison above shows that there must be residues on food block 2 that correspond to the damage on the incoming goods packaging 3. In the example shown, these are the packaging residue 4' resulting from the tear at defect 4, and the surface residue 5' resulting from the defect when the waxed inner layer of the incoming goods packaging 3 was removed. Data analysis 7, by examining the image of the incoming goods packaging 3, has thus determined that the incoming product for food production is contaminated by residues from the incoming goods packaging 3 and can automatically ensure that the affected food block 2 undergoes manual inspection or other processing before further processing.
[0043] The logistical tasks associated with carrying out the procedure, which in Fig.The process shown schematically as an example can be carried out using standard manufacturing equipment. For instance, two conveyor belts can feed the frozen and packaged food blocks 1 into the processing plant, while automated unpacking devices known from the prior art or workstations for personnel can open the incoming goods packaging 3 and place the unpacked food block 2 on one side and the unfolded incoming goods packaging 3 on the other onto separate conveyor belts.
[0044] The incoming goods packaging is fed to the optical detection devices manually or automatically. The method used will depend on the intended application, the quantities of products to be processed, and other factors. For example, the feeding can be carried out via a conveyor belt for the incoming goods packaging 3, which may include or pass through camera or scanning devices, enabling automatic scanning of the relevant surfaces that have come into contact with the food block 2. The control of the conveyor belts, any necessary transfer stations, or other means for transferring, for example, the food block 2 to another transport device or for placing the unfolded incoming goods packaging 3, is carried out by software embedded in the data evaluation unit 7 or a separate control system.
[0045] The allocation of the food blocks 2 to the now separated, disassembled incoming goods packaging 3 can be handled automatically by the data evaluation system 7 using optical markings, workpiece carriers, or by recording the sequence of the parts lying on the respective conveyor belts. This ensures that even after the incoming goods packaging 3 has been separated from the food block 2, a clear identification of the food block 2 assigned to a damaged incoming goods packaging 3 is possible. These measures are well-known from industrial mass production. Reference symbol list: 1 Packaged Food Block 2 Food block 3 Incoming goods packaging 4. Missing section due to demolition 4' Tear off packaging 5. Defect caused by the detachment of a surface layer 5' Surface residue 6 scanning tools 7 Data analysis QUOTES INCLUDED IN THE DESCRIPTION
[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature
[0000] WO 2018 / 236 844 A1
[0006] EP 3 801 934 B1
[0006] EP 3 895 541 B1
[0007]
Claims
[1] Method for introducing food blocks (2) into a food processing plant in which the food blocks (2) are further processed as food, wherein the process is supplied with deep-frozen food blocks (1) packaged in an input product packaging (3) and the deep-frozen food blocks (2) are unpacked by removing the input product packaging (3) and subsequently further processed, wherein after removal of the input product packaging (3) a quality control is carried out using scanning means (6) and an electronic data evaluation (7), characterized by, that the quality control includes the optical scanning of the incoming goods packaging (3) after its removal from the frozen food block (2) using the scanning means (6) and a comparison of the result of the optical scanning with at least one target image formed from an image or other data suitable for a target-actual comparison, in particular in the form of a pixel threshold as a target value in pixels, which corresponds to the data of an undamaged, complete incoming goods packaging (3), wherein, as a result of the comparison via the data evaluation (7), the conformity of the result of the optical scanning with the target image is assessed as non-critical and a deviation therefrom as critical, and the food block (2) is supplied to further processing if the result is non-critical and is supplied to an additional inspection to check for possible remnants of the incoming goods packaging (3) if the result is critical. [2] Method for introducing food blocks (2) into a food processing plant according to claim 1, characterized by , that the quality control additionally includes the optical detection of the deep-frozen food block (2) after removal of the incoming goods packaging (3) by means of the scanning means (6) and a comparison of the result of the optical detection with at least one target image that corresponds to the image of a food block (2) without packaging residues (4') or surface residues (5') of the incoming goods packaging (3), [3] Method for introducing food blocks (2) into a food processing plant according to claim 1 or claim 2, characterized bythat the incoming goods packaging (3) contains a mark which can be detected by the scanning means (6), in particular by illuminating the incoming goods packaging (3) after removal from the food block (2) or by illuminating the food block (2) after removal of the incoming goods packaging (3) by means of a light of a defined wavelength, in particular ultraviolet light. [4] Method for introducing food blocks (2) into food processing according to claim 3, characterized by, that the incoming goods packaging (2) is multi-layered, wherein, as a security feature, an inner layer from the perspective of the packed food block (2) is provided with a coating, an insert behind an at least partially transparent cover or a print which changes its appearance under irradiation with ultraviolet light, becoming transparent, invisible or visible, and the data evaluation (7) uses the detection of the security feature to assess the comparison. [5] Method for introducing food blocks (2) into a food processing plant according to any of the preceding claims, characterized by, that the incoming goods packaging (3) and the food block (1) to be unpacked, still packaged, or the food block (2) already unpacked, are provided with a unique identifier directly or indirectly via an assigned workpiece carrier and / or that the sequence of the food blocks (2) and empty incoming goods packaging (3) flowing out of the packing station enables the identification of an incoming goods packaging to a specific food block (2) and the identification of the food block (2) in question by recording the sequence of the food blocks (2) and the empty incoming goods packaging flowing out of the packing station. [6] Method for introducing food blocks (2) into a food processing plant according to any of the preceding claims, characterized by , that the packaging of incoming goods (3) shall be paper, cardboard or film packaging provided with a barrier layer, in particular packaging made of waxed paper or cardboard. [7] Method for introducing food blocks (2) into a food processing plant according to any of the preceding claims, characterized by , that cubic or cuboid food blocks (2) are used, wherein, in the event of a critical result being detected, the possible position of any residue of the incoming goods packaging (3) adhering to the food block (2), in particular in the form of a surface residue (5') or a packaging tear (4'), is extracted from the result of the optical detection and the comparison with the target image, and the side of the food block (2) on which the position is located is defined and / or marked as the critical side. [8] Method for introducing food blocks (2) into a food processing plant according to the preceding claim, characterized by, that a layer is cut off from the critical side, whereby the cut-off layer is either disposed of or subjected to additional inspection with the aim of further processing the cut-off food material. [9] Method for introducing food blocks (2) into a food processing plant according to the preceding claim, characterized by , that the weight of the section is determined by means of optical detection and evaluation or by means of a scale and is taken into account in the further processing of the food block (2) into food portions with a required unit weight or that the thickness of the layer is determined in such a way that the food block (2) can be divided into integer multiples of the unit weight, possibly with a required geometry, in the subsequent processing. [10] Method for introducing food blocks (2) into a food processing plant according to any one of the preceding claims, characterized by , that after a critical result is detected, the affected food block (2) is checked manually or via optical detection means for any remaining parts of the incoming goods packaging (3) as part of the additional control, whereby any remaining parts found are removed and optically detected and compared with the target image in order to determine whether the differences that occurred when comparing the result of the optical detection via the scanning means (6) with the target image are compensated by adding the detected remaining parts of the incoming goods packaging (3), whereby if the difference is compensated the result of the comparison is set to non-critical and the food block (2) is sent for further food processing. [11] Method for introducing food blocks into a food processing plant according to any of the preceding claims, characterized by, that the scanning means (6) are formed by at least one flatbed scanner or other devices and apparatus for generating an at least two-dimensional image, which is preferably able to scan the front and back of the incoming goods packaging (3) which is folded into a flat position after removal from the food block (2), wherein for this purpose the incoming goods packaging (3) is folded in such a way that all surfaces which have been in contact with the food block (2) are arranged on the underside and / or the top side and the scanning means (6) are designed in such a way that they are able to detect these surfaces. [12] Method for introducing food blocks into a food processing plant according to any of the preceding claims, characterized by , that the scanning means (6) shall include a camera, an infrared camera or other optical image capture means. [13] Method for introducing food blocks into a food processing plant according to any of the preceding claims, characterized by , that the incoming goods packaging (3) has an inner layer, in particular a waxed layer and an outer layer, wherein the inner layer has a different color than the outer layer for better detectability via the scanning means (6). [14] Method for introducing food blocks into a food processing plant according to any of the preceding claims, characterized by , that (2) food blocks shall be made of meat, game, fish, vegetables, protein-containing meat substitutes or mixtures thereof. [15] Processing line or apparatus for carrying out a method according to any one of claims 1-14, characterized by , that the processing line or device has: • a feeder for frozen, packaged food blocks (1) to an unpacking station in which the packaged food blocks (1) are separated manually or mechanically from an incoming goods package (3), • a scanning device with scanning means (6) for optically detecting at least the surfaces of the incoming goods packaging (3) that have been in contact with the deep-frozen, packaged food block (1), • a data evaluation system (7) associated with the scanning device in the form of a data processing system for carrying out a target-actual comparison, within the framework of which the data evaluation system (7) is able to determine a difference between a stored target image or a target image already determined by evaluation of the optically captured appearance of previous optical capture by the scanning device with the result of the optical capture captured by the scanning device and • a device for diverting food blocks (2) whose incoming goods packaging (3) was classified as critical in the target-actual comparison from the processing line or the device.
Citation Information
Patent Citations
Process and system for in-line inspection of product stream for detection of foreign objects
EP3801934B1
Method of providing control for a food processing system and a sensor device therefor
EP3895541B1
Quality control method for a machine for producing packaging
US20130177233A1
Method of inspecting the quality of blanks, in particular of blanks to be processed into packaging material, and quality inspection system
US20190147576A1
System and method for hyperspectral image processing to identify object
WO2018236844A1