Detection of Heat-Treated Marks on Wooden Pallets

Abstract

The pallet inspection system includes a frame configured to have a pallet receiving area for receiving a wooden pallet to be inspected for having at least one mark indicative of the wood of the pallet being heat treated. A camera is carried by the frame and generates an image of the wooden pallet in response to the wooden pallet being in the pallet receiving area. A processor performs object detection on each image to detect whether the mark is present, crops each image having the mark to remove areas surrounding the mark in the image, and performs image segmentation on each cropped image to classify pixels in the cropped image into regions. The processor determines the readability of the regions in each cropped image based on a respective readability criterion threshold. The mark in each cropped image is classified as readable based on the mark satisfying the respective readability criterion threshold.

Description

Technical Field

[0001] The present disclosure relates to pallets, and more particularly to detecting marks on wooden pallets indicating that the wood within the pallet has been heat treated.

Background Art

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 262,453, filed Oct. 13, 2021, which is hereby incorporated by reference in its entirety.

[0003] Wooden pallets are used to transport various bulk goods and equipment required in manufacturing and warehouse operations. Wooden pallets used in international transportation are heat treated. Heat treatment is one of the steps necessary to ensure the safety of the environment at the destination of the products being transported, as well as the products themselves.

[0004] Wooden pallets are made of organic materials. Wood does not grow in a sterile environment. The ground where the wood takes root, the air that surrounds the wood, and the water that the wood absorbs are not only filled with nutritious substances but also have many pests. These pests, whether mature or in the larval stage, unfortunately, are carried too easily from one region of the world to another while remaining attached to the pallet material. Therefore, governments, environmentalists, and pallet manufacturers have come up with various ways to kill them at the pallet manufacturing stage so that they are not brought into places where they are not native or do not belong.

[0005] The heat treatment of pallets is a phytosanitary process developed by the International Plant Protection Convention (IPPC), approved by the World Trade Organization, and supervised by the Food and Agriculture Organization. Its purpose is to prevent and control the entry and spread of pests and plant products. When a pallet is heat treated, a globally recognized image enabling more efficient transportation of goods is stamped or marked.

Summary of the Invention

Means for Solving the Problem

[0006] The pallet inspection system includes a rectangular frame configured to have a pallet receiving area that receives a wooden pallet to be inspected because the wooden pallet has at least one mark indicating that the wooden pallet has been heat-treated. A plurality of cameras are carried by the frame in response to the wooden pallet being in the pallet receiving area and generate an image of the wooden pallet.

[0007] A processor is coupled to the plurality of cameras and is configured to receive the images for processing. The processor includes performing object detection on each image to detect whether the mark exists, trimming each image having the mark to remove an area surrounding the mark within the image, and performing image segmentation on each trimmed image to classify pixels within the trimmed image into regions.

[0008] The readability of the regions in each trimmed image is determined based on respective readability criterion thresholds. The mark in each trimmed image is classified as readable based on the mark satisfying the respective readability criterion thresholds.

[0009] The classified regions for each trimmed image may include a boundary region, a symbol region, and an alphanumeric region, and the pixels within each region may have respective classification identifiers associated therewith.

[0010] The boundary region may have a rectangular shape having a first and a second opposing side and a dividing line extending between one of the opposing sides. The symbol region and the alphanumeric region may be surrounded by the boundary region and separated by the dividing line.

[0011] The classified region may include a boundary region having a classification identifier associated therewith. Determining the readability of the boundary region may include performing corner detection to detect corner points, sampling pixels between the detected corner points, and determining the number of the sampled pixels having the same classification identifier. The boundary region is identified as readable based on the determined number of the sampled pixels having the same classification identifier that exceeds a boundary region threshold.

[0012] The classified region may include a symbol region having a classification identifier associated therewith. Determining the readability of the symbol region may include sampling pixels within the symbol region and determining the number of the sampled pixels having the same classification identifier. The symbol region is identified as readable based on the determined number of the sampled pixels having the same classification identifier that exceeds a symbol region threshold.

[0013] The classified region may include an alphanumeric region having a classification identifier associated therewith. Determining the readability of the alphanumeric region may include identifying pixels within the alphanumeric region having the same classification identifier and determining a readability score of the identified pixels. The readability score may be selected within a range of readability scoring. The alphanumeric region is identified as readable based on the readability score that exceeds a readability score threshold.

[0014] The classified region may include an alphanumeric region having alphanumerics. The processor may be further configured to perform the following for each mark classified as readable. Detecting lines within the alphanumeric region, each line including alphanumerics. Performing optical character recognition to read the alphanumerics of each line.

[0015] The processor may be further configured to perform the following in response to the wooden pallet having a pair of marks each classified as readable. Comparing an alphanumeric read within one of the marks with an alphanumeric read with the other mark. Classifying that the wooden pallet complies in response to each alphanumeric within each mark matching.

[0016] The camera may be arranged such that each side of the pallet receiving area has a single camera focused on a part of a side view of the wooden pallet where the mark is expected to be placed.

[0017] The camera may have a pair of cameras on each side of the pallet receiving area, and the pair of cameras may be arranged to provide overlapping images of all side views of the wooden pallet.

[0018] Another aspect is directed to a method for detecting heat-treated marks on a wooden pallet using the pallet inspection system described above. The method includes generating an image of the wooden pallet, performing object detection on each image to detect whether a mark is present, and trimming each image having the mark to remove an area surrounding the mark within the image. Image segmentation is performed on each trimmed image, and pixels within the trimmed image are classified into regions. Based on respective readability criterion thresholds, the readability of the regions within each trimmed image is determined. Based on the marks in each trimmed image meeting their respective readability criterion thresholds, the marks are classified as readable.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13A

Figure 13B

Figure 14A

Figure 14B

Figure 15A

Figure 15B

Figure 16A

Figure 16B

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

DETAILED DESCRIPTION OF THE INVENTION

[0020] This specification is made with reference to the accompanying drawings that illustrate exemplary embodiments. However, many different embodiments may be used, and thus this specification should not be construed as limited to the particular embodiments described herein. These embodiments are rather provided so that this disclosure will be thorough and complete. Like numbers refer to like elements throughout.

[0021] Shipping products using wood packaging between countries is a process regulated by the International Plant Protection Convention (IPPC). Wooden materials such as pallets can carry diseases and insects from one country to another, and if they spread, they can have an adverse impact on the ecosystem. The IPPC, which consists of more than 183 member countries, has established requirements regarding the handling of wood packaging for import and export in order to prevent the entry of pests that may be harmful to the local plant life.

[0022] According to International Standard for Phytosanitary Measures No. 15 (ISPM15), wood materials exceeding 6 mm in width require debarking and heat treatment or methyl bromide fumigation. When heat-treated, wooden pallets need to be treated for at least 30 minutes and maintain a core temperature of 133°F. After heat treatment or fumigation, the wooden pallets must be stamped with a compliance mark or branded.

[0023] To indicate appropriate heat treatment or methyl bromide fumigation for wooden pallets, a 2-inch stamp or compliance mark 20 is required, and an exemplary format thereof is shown in FIG. 1. The illustrated format of the compliance mark 20 occupies two lines. This format is not restrictive. For example, the compliance stamp or mark 20 may have a format that occupies a single line or more than two lines. The compliance stamp or mark 20 may also be referred to as an ISPM15 mark, a heat-treated marking, or a mark. An exemplary image of the ISPM15 mark 20 on the area of a wooden pallet is shown in FIG. 2.

[0024] The ISPM15 mark 20 includes an outer perimeter 22 and a dividing line 24. The outer perimeter 22 is rectangular, and the dividing line 24 extends between one of the opposing sides of the outer perimeter 22. The outer perimeter 22 and the dividing line 24 are referred to as a boundary region 25.

[0025] In the region to the left of the dividing line 24, there is an IPPC certification symbol. The IPPC certification symbol includes a tree symbol 28 having the letters IPPC 30 adjacent to the tree symbol 28. This region is referred to as a symbol region 29.

[0026] In the region to the right of the dividing line 24, there are alphanumerics. This region is referred to as an alphanumeric region 31. The alphanumerics include a country code 32, a producer code 34, and a treatment code 36. The country code 32 is two characters. By way of example, ES represents Spain, US represents the United States, GB represents the United Kingdom, and AU represents Australia. The producer code 34 is a series of unique alphanumerics indicating the wood treatment agent or packaging manufacturer. This is a unique authentication number that ensures that the wood packaging material can be traced back to the wood treatment agent or packaging manufacturer. The treatment code 36 represents the treatment applied to the wood packaging material. HT is the code for heat treatment, and MB is the code for methyl bromide fumigation.

[0027] The ISPM15 mark 20 is typically required every 24 inches along a wooden pallet. Non - compliance can result in the shipment being rejected by customs, which can lead to expensive fees associated with the re - export of goods for the importer.

[0028] Therefore, it is necessary to automate the detection of the ISPM15 mark 20 on wooden pallets. This is particularly required in high - volume production industries where pallet pooling keeps the industry - wide costs lower than one - way pallets.

[0029] After the bulk goods and equipment are unloaded from the stacked pallets, the wooden pallets are returned to the pallet inspection and repair facility. As part of the inspection, the ISPM15 mark 20 is detected. For the wooden pallet 40 to be compliant, a pair of ISPM15 marks 20 are identified and the alphanumeric characters within each alphanumeric area 31 must match each other.

[0030] Referring now to FIG. 3, an exploded view of an exemplary wooden pallet 40 is illustrated. The wooden pallet 40 is for purposes of explaining different placement positions of the ISPM15 mark 20. The illustrated wooden pallet 40 is not limited as other wooden pallet configurations are readily available.

[0031] The wooden pallet 40 includes an upper deck 50, a lower deck 60, and a plurality of wooden support blocks 70, 72 coupled between the upper deck and the lower deck. The support blocks 7 0, 72 form a gap between the upper deck 50 and the lower deck 60 to receive a lifting member such as the tip of a forklift.

[0032] The upper deck 50 includes a pair of wooden end deck boards 52 spaced apart and a wooden intermediate deck board 54 disposed between the end deck boards 52. Also included within the upper deck 50 are a pair of wooden connector boards 56 spaced apart and a wooden intermediate connector board 58. The connector board 56 and the intermediate connector board 58 are orthogonal to the end deck boards 52 and the intermediate deck board 54. The end deck boards 52 and the intermediate deck board 54 are disposed on the connector board 56 and are directly coupled to the support blocks 70, 72 via nails.

[0033] The lower deck 60 includes lower deck boards 62, 64 oriented in the same direction as the end deck boards 52 and the intermediate deck board 54 within the upper deck 50. The lower deck boards 62, 64 may also be referred to as base boards and are directly coupled to the support blocks 70, 72 via nails.

[0034] The support blocks include corner support blocks 70 and central support blocks 72 between the corner support blocks 70. In total, there are nine support blocks 70, 72 arranged in three columns. Each of the outer columns includes a pair of outer support blocks 70 and a single central support block 72, and the central column includes all the central support blocks 72. The corner support blocks 70 and the central support blocks 72 each have a rectangular shape.

[0035] The different placement positions of the ISPM15 mark 20 include, for example, the central support block 72 as indicated by the squared numbers 1 and 2, the outer lower deck board 62 as indicated by the squared number 3, and the outer edge of the end connector board 56 in the upper deck 50 as indicated by the squared number 4.

[0036] A side view of the stacked wooden pallet 40 having the ISPM15 mark 20 on the central support block 72 is shown in FIG. 4. An upper perspective view of the stacked wooden pallet 40 having the ISPM15 mark 20 on the outer lower deck board 62 is shown in FIG. 5. A side view of the wooden pallet 40 having the ISPM15 mark 20 at the end of the connector board 56 of the upper deck 50 is shown in FIG. 6. When the ISPM15 mark 20 is at the end of the connector board 56, the alphanumeric characters are arranged so as to fit on a single line.

[0037] Next, referring to FIGS. 7-9, a fully enclosed pallet inspection station 100 is configured to inspect the ISPM15 mark 20 of the wooden pallet 40. The pallet inspection station 100 includes a frame 102 having upper and lower covers 104, 106. The frame 102 has a rectangular shape and a pallet receiving area 110 as shown in FIG. 9. Although not shown, a conveyor may be used to convey the wooden pallet 40 through the pallet inspection station 100 for inspection.

[0038] Referring to FIGS. 10 and 11 here, the upper cover 104 is removed, and the camera 120 and the pallet receiving area 110 are exposed. In FIG. 12, both the upper and lower covers 104, 106 are removed, and the camera 120 and the pallet receiving area 110 are exposed. The camera 120 is attached to the frame 102 via a camera arm extension 122. Each camera arm extension 122 extends outwardly upward from the frame 102 such that the camera 120 looks down on the wooden pallet 40 within the pallet receiving area 110. The camera 120 is disposed within a range of 20 degrees to 45 degrees with respect to the wooden pallet 40 within the pallet receiving area 110. Lights are also mounted on each side of the frame 102 and are arranged to illuminate each side of the pallet 40 for each respective camera 120.

[0039] In the illustrated embodiment, there are a total of eight cameras 120, with two cameras 120 on each side. The cameras 120 may be color or monochrome. In another embodiment, there may be a total of four cameras 120, with one camera 120 on each side. In yet another embodiment, there may be two or more cameras 120 on each side.

[0040] The eight cameras 120 are all triggered simultaneously to generate an image of the wooden pallet 40. When the wooden pallet 40 reaches the pallet receiving area 110, the movement of the wooden pallet 40 is stopped. After the wooden pallet 40 is stopped by a stopper within the path of the pallet, the eight cameras 120 are triggered.

[0041] By having two cameras 120 on each side of the frame 102, all or complete side views of the wooden pallet 40 are obtained for processing. Since a single camera 120 cannot provide all or complete side views of the wooden pallet 40, the two cameras 120 on each side provide overlapping images.

[0042] As described above with reference to FIG. 3, the ISPM15 mark 20 can be at different positions on the wooden pallet 40. For example, these positions include the central support block 72 as indicated by the boxed numbers 1 and 2, the outer lower deck board 62 as indicated by the boxed number 3, and the outer edge of the connector board 56 in the upper deck 50 as indicated by the boxed number 4.

[0043] Although not shown, the ISPM15 mark 20 can be on the corner support block 70, the connector board 56, and the intermediate connector board 58. The lower deck boards 62, 64 can also have the ISPM15 mark 20. In yet another example, the ISPM15 mark 20 can be on the upper surface of any of the boards within the upper deck 50, or on the bottom surface of any of the boards within the lower deck 60.

[0044] However, if the wooden pallet 40 being inspected has the ISPM15 mark 20 at the same position on each pallet, one camera 120 may be used for each side. In this case, each camera 120 is focused or positioned to view the same position on the side of the wooden pallet 40.

[0045] Since there are eight cameras 120, eight images are generated for the inspection of the wooden pallet 40. Partial end views of one side of the wooden pallet 40 being inspected are provided in FIGS. 13A and 13B. This is the back end of the wooden pallet 40 as it passes through the pallet inspection station 100 on the conveyor. Image 150(1) in FIG. 13A includes the left outer support block 70 and the central support block 72. Image 150(2) in FIG. 13B includes the right outer support block 70 and the central support block 72. Images 150(1), 150(2) collectively provide a full side view of the wooden pallet 40 with the images overlapping at the central support block 72.

[0046] Opposite partial end views of the wooden pallet 40 are provided in FIGS. 14A and 14B. This is the front end of the wooden pallet 40 as it passes through the pallet inspection station 100 on the conveyor. The stopper 130 is used to stop the wooden pallet 40 before the camera 120 activates. Image 150(3) in FIG. 14A includes the left outer support block 70 and the central support block 72. In this image 150(3), the ISPM15 mark 20 is on the outer lower deck board 62. Image 150(4) in FIG. 14B includes the right outer support block 70 and the central support block 72.

[0047] Partial side views of the left side of the wooden pallet 40 are provided in FIGS. 15A and 15B. Image 150(5) in FIG. 15A includes the left outer support block 70 and the central support block 72. Image 150(6) in FIG. 15B includes the right outer support block 70 and the central support block 72.

[0048] Partial side views of the right side of the wooden pallet 40 are provided in FIGS. 16A and 16B. Image 150(7) in FIG. 16A includes the left outer support block 70 and the central support block 72. In this image 150(7), the ISPM15 mark 20 is on the outer lower deck board 62. Image 150(8) in FIG. 16B includes the right outer support block 70 and the central support block 72. Images 150(1) - 150(8) are generally referred to as image 150 hereinafter.

[0049] Now, the operation of the pallet inspection station 100 for detecting the ISPM15 mark 20 will be described. A block diagram of the pallet inspection system 95 equipped with the pallet inspection station 100 is provided in FIG. 17, and a flow diagram 200 for detecting the ISPM15 mark 20 on the wooden pallet 40 using the pallet inspection system 95 is provided in FIG. 18.

[0050] The conveyor 105 moves the wooden pallet 40 in the direction of the illustrated arrow through the pallet inspection station 100. The conveyor 105 includes a sensor 132 at the entrance of the pallet inspection station 100 to detect the arrival of the wooden pallet 40. The sensor 132 is coupled to a controller 134.

[0051] The sensor 132 can be configured as, for example, a photoelectric sensor. The photoelectric sensor includes a transmitter and a receiver on opposite sides of the conveyor 105. The transmitter sends an optical signal, which can be visible or infrared, to the receiver. The wooden pallet 40 is detected when the light beam is blocked from reaching the receiver from the transmitter.

[0052] When the wooden pallet 40 arrives, the controller 134 activates a stopper 130 within the path of the wooden pallet 40. When activated, the stopper 130 is lifted through a gap in the conveyor 105 and stops the pallet at a position set relative to the camera 120. After the wooden pallet 40 is stopped by the stopper 13 0 The controller 134 then activates or triggers the camera 120 to generate an image 150 of the wooden pallet 40.

[0053] The image 150 is sent to a processing unit 140 for processing. The processing unit 140 executes different machine learning algorithms, as will be described in more detail below. The processing unit 140 can be, for example, a graphics processing unit (GPU), a central processing unit (CPU), or an edge computing device.

[0054] In the flowchart 200, the generated image 150 of the inspected wooden pallet 40 is received at block 202. The GPU 140 executes an object detection algorithm 150 trained to locate the ISPM 15 mark 20 within the image 150 at block 204.

[0055] The object detection algorithm 150 can operate based on artificial intelligence (AI) and machine learning (ML) to determine the ISPM15 mark 20 within the image 150. The object detection algorithm 150 is trained using annotated images that include different positions where the ISPM15 mark 20 can be placed. In the annotated images, bounding boxes are used to mark different positions of the ISPM15 mark 20.

[0056] In other embodiments, a segmentation algorithm can be used instead of the object detection algorithm 150. The segmentation algorithm divides an image into a set of pixels or regions. The purpose of the division is to better understand what the image represents. The set of pixels can represent objects within the image that are of interest for specific applications such as the detection of the ISPM15 mark 20. Instead of object detection, direct segmentation can be used to trim the image processed by the segmentation algorithm.

[0057] In block 204, if the image 150 does not have the ISPM15 mark 20, the image is discarded in block 206. If the image 150 has the ISPM15 mark 20, the image 150 is trimmed in block 208. In the trimmed image 250, as shown in FIG. 19, the ISPM15 mark 20 is trimmed such that the area surrounding the ISPM15 mark 20 within the image 150 is removed.

[0058] Next, the trimmed image 250 is passed to the pixel segmentation algorithm 152 in block 210. Image segmentation is a process of classifying or assigning labels to all the pixels within the trimmed image 250 such that pixels having the same classification identifier share specific characteristics. The pixel segmentation algorithm 152 can operate based on artificial intelligence (AI) and machine learning (ML).

[0059] The trimmed image 250 is segmented into a border region 25, a symbol region 29, and an alphanumeric region 31 as described above and shown in FIG. 1. The pixels within the border region 25 may have the number 1 assigned as a classification identifier. The pixels within the symbol region 29 may have the number 2 assigned as a classification identifier. The pixels within the alphanumeric region 31 may have the number 3 assigned as a classification identifier. A fourth region is the background region of the pixels outside the border region 25, the symbol region 29, and the alphanumeric region 31. The pixels within the background region may have the number 4 assigned as a classification identifier.

[0060] The output of the pixel segmentation algorithm 152 is provided to respective readability algorithms 154. Each readability algorithm 154 may operate based on artificial intelligence (AI) and machine learning (ML). The readability algorithm 154 analyzes the regions based on respective readability criteria associated with each region. The readability criteria are used to determine whether each region is legible enough to be read and understood by a person. The readability algorithm 154 does not read the regions.

[0061] The readability algorithm 154 includes a first readability algorithm 154(1) for the border region 25, a second readability algorithm 154(2) for the symbol region 29, and a third readability algorithm 31 154(3) for the alphanumeric region. The readability algorithms 154 are executed simultaneously by the processor. That is, the different regions are analyzed simultaneously by their respective readability algorithms 154.

[0062] The first readability algorithm 154(1) is used to analyze the boundary region 25 in block 212. The first readability algorithm 154(1) is trained to perform corner detection to detect corner points, as shown in FIG. 20. Corner points 252 are detected with respect to the outer perimeter 22, and corner points 254 are detected with respect to the dividing line 24. Pixels between the detected corner points 252, 254 are sampled, and the number of sampled pixels having the same classification identifier 1 is determined.

[0063] The boundary region 25 is identified as readable in block 214 based on the determined number of sampled pixels having the same classification identifier 1 that exceed the boundary region threshold. The boundary region threshold includes a threshold for the outer perimeter 22 and a threshold for the dividing line 24.

[0064] Each threshold corresponds to the percentage of sampled pixels to be resent. For example, the threshold for the outer perimeter 22 can be in the range of 70% to 100%, and the threshold for the dividing line 24 can be in the range of 95% to 100%. If the boundary region 25 is not readable, the image 150 is discarded at block 226. If the boundary region 25 is readable, the process proceeds to block 224.

[0065] The second readability algorithm 154(2) is used to analyze the symbol region 29 in block 218. The second readability algorithm 154(2) is trained to analyze the pixels within the tree 28 and the pixels forming the IPPC characters 30 adjacent to the tree, as shown by the image 260 in FIG. 21. The pixels within the tree 28 are sampled, and the pixels within the IPPC characters 30 are sampled. The number of sampled pixels having the same classification identifier 2 is determined.

[0066] The symbol region 29 is identified as readable in block 218 based on the determined number of sampled pixels having the same classification identifier 2 that exceed the symbol region threshold. The symbol region threshold includes a threshold for the tree 28 and a threshold for the IPPC characters 30.

[0067] Each threshold corresponds to a percentage of the existing sampled pixels. For example, the threshold for tree 28 can be in the range of 75% - 100%, and the threshold for IPPC character 30 can also be in the range of 75% - 100%. The readability criteria associated with IPPC character 30 may be such that IPPC character 30 is visible but not necessarily readable. If only one of the IPPC characters is not visible, IPPC character 30 is considered readable. If symbol area 29 is not readable, in block 226, image 150 is discarded. If symbol area 29 is readable, the process proceeds to block 224.

[0068] The third readability algorithm 154(3) is used to analyze alphanumeric area 31 at block 220. The third readability algorithm 154(3) is trained to analyze the pixels within the alphanumeric characters in alphanumeric area 31. Pixels within alphanumeric area 31 having the same classification identifier 3 are identified.

[0069] A readability score is determined for the identified pixels, and the readability score is selected within a readability scoring range. The readability scoring range can vary, for example, between 1 and 5. 5 can correspond to all alphanumeric characters that are readable, as shown by image 262 in FIG. 22. 1 can correspond to most of the visible alphanumeric characters, as shown by image 264 in FIG. 23.

[0070] 4 can correspond to one of the alphanumeric characters that is partially visible but the alphanumeric character is still known. 3 can correspond to one of the alphanumeric characters that is not displayed or missing, and 2 can correspond to two or more alphanumeric characters that are not displayed or missing.

[0071] The use of the readability scoring range provides flexibility to a third readability algorithm 154(3) when determining alphanumeric readability. Instead of the determination being binary as in the case of using the first and second readability algorithms 154(1), 154(2), the third readability algorithm 154(3) allows for flexibility in making the determination. When the determination is at the center of the readability scoring range (i.e., 2 to 4), the slide scale allows for a more general determination to be made about alphanumeric readability.

[0072] The alphanumeric region 31 is identified in block 222 as readable based on a readability score exceeding a readability score threshold. The readability score threshold can be, for example, 3.5. If the alphanumeric region 31 is not readable, the image 150 is discarded in block 226. If the alphanumeric region 31 is readable, the process proceeds to block 224.

[0073] As an alternative to using the second readability algorithm 154(2) to analyze the symbol region 29, the third readability algorithm 154(3) can be configured to analyze the symbol region 29. That is, the readability criteria for the symbol region 29 will be based on a readability scale similar to the readability scale described for the alphanumeric region 31.

[0074] For the ISPM15 mark 20 to be readable, each of the border region 25, the symbol region 29, and the alphanumeric region 31 needs to be readable. If one of the three regions is unreadable, the ISPM15 mark 20 is classified as unreadable in block 226. If all three regions are readable, the ISPM15 mark 20 is classified as readable and the process proceeds to block 228.

[0075] After the ISPM15 mark 20 is identified as readable in the received image 150, the next step in the process is to determine whether the wooden pallet 40 is compliant. This determination is based on the form. For the wooden pallet 40 to be compliant, a pair that matches the ISPM15 mark 20 is required. If there is only one ISPM15 mark 20, or if the alphanumerics within the two ISPM15 marks 20 do not match each other, the wooden pallet 40 is classified as non-compliant.

[0076] The compliant pallet algorithm 156 is used to determine whether the wooden pallet 40 has a pair of matching ISPM15 marks 20. The compliant pallet algorithm 156 first detects the line 270 within the alphanumeric region 31 for each ISPM15 mark 20 at block 230, as shown by the image 266 in FIG. 24. There are three lines 270 in the alphanumeric region 31, and each line 270 contains alphanumerics.

[0077] After the line 270 is detected, next, as shown by the display 268 in FIG. 25, optical character recognition (OCR) is performed at block 232 to read the alphanumerics of each line. At block 234, a determination is made as to whether the alphanumerics within the pair of ISPM15 marks 20 match each other.

[0078] If the alphanumerics match, the wooden pallet 40 is classified as compliant at block 236. If the alphanumerics do not match, the wooden pallet 40 is classified as non-compliant at block 238. In other embodiments of the processing unit 140 that receive images for processing, the processing unit 140 may not be able to trim each image. Instead, the mark is detected using object detection, and then the readability is determined to classify the mark. Based on the marks that meet their respective readability criteria thresholds, the marks are classified.

[0079] Another aspect is directed to a method for operating the pallet inspection system 95 as described above. Here, FIG. 2 6Referring to the flowchart 300 of FIG. 3, starting from the start (block 302), the method includes generating an image 150 of the wooden pallet 40 at block 304 and performing object detection on each image 150 at block 306 to detect whether the ISPM 15 mark 20 is present. Each image 150 having the ISPM 15 mark 20 is trimmed at block 308, and the area surrounding the mark in the image is removed.

[0080] Image segmentation is performed on each trimmed image 250 at block 310, and the pixels in the trimmed image 250 are classified into regions. The readability of the regions in each trimmed image 250 is determined at block 312 based on their respective readability criterion thresholds. The ISPM 15 mark 20 in each trimmed image 250 is classified as readable at block 314 based on the ISPM 15 mark 20 meeting their respective readability criterion thresholds. The method ends at block 316.

[0081] Those skilled in the art having the benefit of the teachings presented in the above description and associated drawings will envision many modifications and other embodiments. Accordingly, the above is not to be limited to the exemplary embodiments, and it will be understood that modifications and other embodiments are included in the appended claims.

Claims

1. A rectangular frame configured to have a pallet receiving area for receiving a wooden pallet to be inspected for having at least one mark indicating that the wood of the pallet has been heat treated, a plurality of cameras carried by the frame in response to the wooden pallet being in the pallet receiving area, the plurality of cameras configured to generate an image of the wooden pallet, a processor coupled to the plurality of cameras and configured to receive the image for processing, the processing comprising: performing object detection on each image to detect whether the mark is present; trimming each image having the mark to remove an area surrounding the mark within the image; performing image segmentation on each trimmed image to classify pixels within the trimmed image into regions; determining the readability of each classified region within each trimmed image based on a readability criterion threshold for each classified region; classifying the mark within each trimmed image as readable based on each classified region having met the respective readability criterion threshold for that classified region; a processor including; a pallet inspection system comprising.

2. The classified regions for each trimmed image include a border region, a symbol region, and an alphanumeric region, and pixels within each region have respective classification identifiers associated therewith. The pallet inspection system according to claim 1.

3. The border region has a rectangular shape having first and second opposing sides and a dividing line extending between one of the opposing sides, and the symbol region and the alphanumeric region are surrounded by the border region and separated by the dividing line. The pallet inspection system according to claim 2.

4. The classified regions include a border region having a classification identifier associated therewith, and determining the readability of the border region comprises: performing corner detection to detect corner points; sampling pixels between the detected corner points; determining the number of sampled pixels having the same classification identifier; identifying the border region as readable based on the determined number of sampled pixels having the same classification identifier exceeding a border region threshold. The pallet inspection system according to claim 1.

5. The classified area includes a symbol area having a classification identifier associated therewith, and determining the readability of the symbol area includes sampling pixels within the symbol area, determining the number of the sampled pixels having the same classification identifier, and identifying the symbol area as readable based on the determined number of the sampled pixels having the same classification identifier that exceeds a symbol area threshold. The pallet inspection system according to claim 1.

6. The classified area includes an alphanumeric area having a classification identifier associated therewith, and determining the readability of the alphanumeric area includes identifying pixels within the alphanumeric area having the same classification identifier, and determining a readability score of the identified pixels, wherein the readability score is selected within a range of readability scoring, and identifying the alphanumeric area as readable based on the readability score that exceeds a readability score threshold. The pallet inspection system according to claim 1.

7. The classified area includes an alphanumeric area having alphanumerics, and the processor, for each mark classified as readable, detects lines within the alphanumeric area, each line including alphanumerics, and performs optical character recognition to read the alphanumerics of each line, and is further configured to perform. The pallet inspection system according to claim 1.

8. The processor, in response to the wooden pallet having a pair of marks each classified as readable, compares alphanumerics read within one of the marks with alphanumerics read in the other mark, and classifies that the wooden pallet complies in response to the alphanumerics within each mark matching, and is further configured to perform. The pallet inspection system according to claim 7.

9. The plurality of cameras are arranged such that each side of the pallet receiving area focuses on a single camera that is part of a side view of the wooden pallet where the mark is expected to be placed. The pallet inspection system according to claim 1.

10. The plurality of cameras are arranged such that each side of the pallet receiving area has a pair of cameras, and the pair of cameras provides overlapping images of all side views of the wooden pallet. The pallet inspection system according to claim 1.

11. A method for detecting a heat-treated mark on a wooden pallet, comprising: generating an image of the wooden pallet; performing object detection on each image to detect whether a mark is present; trimming each image having the mark to remove an area surrounding the mark in the image; performing image segmentation on each trimmed image to classify pixels in the trimmed image into regions; determining the readability of each classified region in each trimmed image based on a readability criterion threshold for each classified region; classifying the mark in each trimmed image as readable based on the mark satisfying the readability criterion threshold for each classified region; A method comprising the above steps.

12. The classified regions for each trimmed image include a boundary region, a symbol region, and an alphanumeric region, and pixels within each region have respective classification identifiers associated therewith. The method according to claim 11.

13. having a rectangular shape with first and second opposing sides and a dividing line extending between one of the opposing sides, wherein the symbol region and the alphanumeric region are surrounded by the boundary region and separated by the dividing line. The method according to claim 12.

14. The classified regions include a boundary region having a classification identifier associated therewith, and determining the readability of the boundary region includes: performing corner point detection to detect corner points; sampling pixels between the detected corner points; determining the number of sampled pixels having the same classification identifier; identifying the boundary region as readable based on the determined number of sampled pixels having the same classification identifier exceeding a boundary region threshold. The method according to claim 11.

15. The classified regions include a symbol region having a classification identifier associated therewith, and determining the readability of the symbol region includes: Sampling the pixels within the symbol area, Determining the number of the sampled pixels having the same classification identifier, Identifying the symbol area as readable based on the determined number of the sampled pixels having the same classification identifier that exceeds a symbol area threshold, The method according to claim 11.

16. The classified area includes an alphanumeric area having a classification identifier associated therewith, and determining the readability of the alphanumeric area comprises Identifying the pixels within the alphanumeric area having the same classification identifier, Determining a readability score of the identified pixels, wherein the readability score is selected within a range of readability scoring, Identifying the alphanumeric area as readable based on the readability score that exceeds a readability score threshold, The method according to claim 11.

17. The classified area includes an alphanumeric area having alphanumerics, and for each mark classified as readable, Detecting a line within the alphanumeric area, wherein each line includes alphanumerics, Further comprising performing optical character recognition to read the alphanumerics of each line, The method according to claim 11.

18. In response to the wooden pallet having a pair of marks each classified as readable, Comparing the alphanumerics read within one of the marks with the alphanumerics read with the other mark, Further comprising classifying that the wooden pallet complies in response to the alphanumerics within each mark matching, The method according to claim 17.

19. The wooden pallet is received in a pallet receiving area, and each side of the pallet receiving area has a single camera focused on a part of a side view of the wooden pallet where the mark is expected to be arranged, The method according to claim 11.

20. The wooden pallet is received in a pallet receiving area, and each side of the pallet receiving area has a pair of cameras, and the pair of cameras provides overlapping images of all side views of the wooden pallet, The method according to claim 11.

Images