Image recognition method, image recognition device, material sorting equipment and storage medium

Abstract

The invention relates to the technical field of image recognition, in particular to an image recognition method, an image recognition device, material sorting equipment and a storage medium. The invention provides an image recognition method. The image recognition method comprises the steps of obtaining a to-be-processed image when materials are conveyed through a conveying belt; determining a first salient point image in the to-be-processed image, and determining first contour information of the material; based on a matching result between the first salient point image and a reference salient point image, target salient point position information corresponding to the to-be-processed image is determined, and the reference salient point image is a salient point spliced image with a transmission band containing multiple rows of salient points; and based on a matching result of the first contour information and the target salient point position information, updating the first contour information to obtain target contour information of the material so as to identify the material through the target contour information. The material recognition process is facilitated, the material sorting efficiency is improved, and the material recognition performance of the material sorting equipment is guaranteed.

Description

Technical Field

[0001] The present disclosure relates to the technical field of image recognition, and in particular to an image recognition method, an image recognition device, a material sorting device, and a computer-readable storage medium. Background Art

[0002] When the material sorting equipment is transmitting materials, in order to ensure the stability of material transmission and prevent the materials from rolling during the transmission process, some evenly distributed convex points are arranged on the conveyor belt to control the relative movement state of the materials on the conveyor belt through multiple convex points.

[0003] In the process of material transmission, the currently transmitted material can generally be determined by performing image recognition processing on the collected images, but the convex points on the conveyor belt may affect the accuracy of material recognition and the recognition results of the materials. For example, in the process of plastic transmission, if the plastic is small in size or too thin, and it adheres to the convex points on the conveyor belt during the transmission process, the convex points are easily mistakenly recognized as part of the plastic during the image recognition processing, thereby affecting the recognition results of the plastic.

[0004] In the related art, a real-time matching method is used to determine the position of convex points in the image, thereby achieving the purpose of improving the accuracy of material recognition. However, using this method to determine the position of convex points will affect the efficiency of material recognition. Summary of the invention

[0005] In order to overcome the problems existing in the related art, an exemplary embodiment of the present disclosure provides an image recognition method, which is applied to material sorting equipment, wherein the material sorting equipment includes a conveyor belt with multiple rows of convex points, and the conveyor belt is used to convey materials. The method includes: obtaining an image to be processed when the material is conveyed by the conveyor belt; determining a first convex point image in the image to be processed, and determining first contour information of the material; based on a matching result between the first convex point image and a reference convex point image, determining target convex point position information corresponding to the image to be processed, wherein the reference convex point image is a convex point splicing image of the conveyor belt containing multiple rows of convex points; based on a matching result between the first contour information and the target convex point position information, updating the first contour information, and obtaining target contour information of the material, so as to identify the material through the target contour information.

[0006] In some embodiments, if the image to be processed is the first frame image, the target salient point position information corresponding to the image to be processed is determined based on the matching result between the first salient point image and the reference salient point image, including: determining the first row interval salient point image corresponding to the image to be processed in the reference salient point image based on the matching result between the first salient point image and the reference salient point image; determining the first reference position information of each salient point image in the first row interval salient point image according to the reference position information of each salient point image in the reference salient point image; determining the target salient point position information corresponding to the image to be processed based on the first reference position information of each salient point image in the first row interval salient point image.

[0007] In some embodiments, if the image to be processed is not the first frame image, the target salient point position information corresponding to the image to be processed is determined based on the matching result between the first salient point image and the reference salient point image, including: obtaining the position of the second row interval salient point image corresponding to the reference salient point image of the previous frame image in the reference salient point image; determining the third row interval salient point image in the reference salient point image corresponding to the image to be processed based on the position of the second row interval salient point image in the reference salient point image and the image acquisition order of the image to be processed; determining the second reference position information of each salient point image in the third row interval salient point image according to the reference position information of each salient point image in the reference salient point image; determining the target salient point position information corresponding to the image to be processed based on the second reference position information of each salient point image in the third row interval salient point image.

[0008] In some embodiments, the process of determining the reference salient point image includes: obtaining multiple consecutive frames of first images when the conveyor belt is running in a no-load state; splicing multiple frames of first images to obtain an intermediate image, and the number of salient point image rows in the intermediate image is greater than the number of salient point rows of the conveyor belt; extracting the reference salient point image corresponding to the multiple rows of salient points from the intermediate image.

[0009] In some embodiments, extracting reference salient point images corresponding to multiple rows of salient points from an intermediate image includes: matching each row of salient point images in the intermediate image with multiple salient point detection templates to obtain a similarity vector for each row of salient point images, wherein the multiple salient point detection templates are determined based on a background image corresponding to multiple rows of salient points, and the salient point detection templates correspond to each column of salient point images; determining a reference salient point image corresponding to multiple rows of salient points in the intermediate image based on the similarity vector of each row of salient point images in the intermediate image; and extracting the reference salient point image.

[0010] In some embodiments, based on the similarity vector of each row of salient point images in the intermediate image, determining reference salient point images corresponding to multiple rows of salient points in the intermediate image includes: determining a target row of salient point images that have the greatest similarity with the first row of salient point images among other rows of salient point images in the intermediate image; and determining reference salient point images corresponding to multiple rows of salient points in the intermediate image according to the position of the target row of salient point images in the intermediate image, the reference salient point images including a row of salient point images from the first row of salient point images to the previous row of salient point images from the target row of salient point images.

[0011] In some embodiments, extracting a reference salient point image corresponding to multiple rows of salient points from an intermediate image includes: acquiring multiple consecutive frames of second images when the conveyor belt is running in a no-load state; sequentially matching each frame of the second image with the intermediate image, determining an intermediate local image in the intermediate image corresponding to the second image, and determining a reference salient point image corresponding to multiple rows of salient points based on a height relationship between two adjacent intermediate local images; and extracting the reference salient point image from the intermediate image.

[0012] In some embodiments, the height of each frame of the second image is the same; each frame of the second image is matched with the intermediate image in sequence, the intermediate local image corresponding to the second image in the intermediate image is determined, and based on the height relationship between two adjacent intermediate local images, the reference salient point image corresponding to the multiple rows of salient points is determined, including: matching the current second image with the intermediate image, determining the first intermediate local image corresponding to the current second image in the intermediate image, and the first position information of the first intermediate local image, the first position information including the first height of the first intermediate local image; determining the second position information of the second intermediate local image, the second intermediate local image is the local image corresponding to the previous second image in the intermediate image, and the second position information includes the second height of the second intermediate local image; if the difference between the second height and the first height is equal to the frame image height, performing the step of matching the next second image with the intermediate image; if the difference between the second height and the first height is less than the frame image height, determining the reference salient point image corresponding to the multiple rows of salient points in the intermediate image based on the acquisition order of the current second image in multiple frames of second images, the first height and the second height.

[0013] In some embodiments, the process of determining the reference position information of each salient point image in the reference salient point image includes: if the current salient point image is not a first-row salient point image, then determining the reference position information of the current salient point image based on the first-row spacing between the row where the current salient point image is located and the previous row; if the current salient point image is a first-row salient point image, then determining the second-row spacing between the last-row salient point image of the reference salient point image and the next first-row salient point image through the intermediate image, and determining the reference position information of the current salient point image based on the second-row spacing.

[0014] In some embodiments, based on the matching result of the first contour information and the target convex point position information, the first contour information is updated to obtain the target contour information of the material, including: if the target convex point position partially overlaps with the first contour information, the contour information in the first contour information that corresponds to the overlapping target convex point position is removed to obtain the target contour information of the material.

[0015] In a second aspect, the present disclosure also provides an image recognition device, which is applied to material sorting equipment. The material sorting equipment includes a conveyor belt with multiple rows of convex points, and the conveyor belt is used to transport materials. The device includes: a first acquisition module, which is used to acquire the image to be processed when the material is transported by the conveyor belt; a first determination module, which is used to determine the first convex point image in the image to be processed, and determine the first contour information of the material; a second determination module, which is used to determine the target convex point position information corresponding to the image to be processed based on the matching result between the first convex point image and the reference convex point image, and the reference convex point image is a convex point splicing image of the conveyor belt containing multiple rows of convex points; an update module, which is used to update the first contour information based on the matching result between the first contour information and the target convex point position information, and obtain the target contour information of the material, so as to identify the material through the target contour information.

[0016] In a third aspect, the present disclosure further provides a material sorting device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the image recognition method provided in any of the above aspects by executing the computer instructions.

[0017] In a fourth aspect, the present disclosure further provides a computer-readable storage medium, which stores the following program, and the program is used to execute the image recognition method provided by any of the above aspects.

[0018] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.

[0019] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: According to the material identification method provided by the present disclosure, by pre-acquiring reference salient point images corresponding to the multiple rows of salient points included in the conveyor belt, the distribution of each salient point image can be clarified, and then after obtaining the first salient point image of the image to be processed, the target salient point position information corresponding to the image to be processed can be quickly determined through the matching result between the first salient point image and the reference salient point image, thereby improving the information determination efficiency, and then combined with the matching result with the first contour information of the material, the interference of the salient point image on the first contour information can be quickly eliminated or reduced, and the recognition quality of the target contour information can be improved, thereby helping to promote the material identification process, improve the material sorting efficiency, and ensure the material identification performance of the material sorting equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present disclosure may be better understood by describing exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, in which:

[0021] Figure 1 This is a schematic diagram of the structure of a material sorting device shown in an exemplary embodiment of the present disclosure;

[0022] Figure 2 A flowchart of an image recognition method according to an exemplary embodiment of the present disclosure is shown;

[0023] Figure 3 A flowchart of another image recognition method shown in an exemplary embodiment of the present disclosure;

[0024] Figure 4 A flowchart of another image recognition method according to an exemplary embodiment of the present disclosure is shown;

[0025] Figure 5 A schematic flow chart of a method for determining a reference salient point image according to an exemplary embodiment of the present disclosure;

[0026] Figure 6 A schematic flow chart of another method for determining a reference salient point image according to an exemplary embodiment of the present disclosure;

[0027] Figure 7 A schematic flow chart of yet another method for determining a reference salient point image according to an exemplary embodiment of the present disclosure;

[0028] Figure 8 A schematic diagram showing an exemplary embodiment of the present disclosure;

[0029] Fig. 9 Another schematic diagram showing an exemplary embodiment of the present disclosure;

[0030] Fig.10 A schematic diagram of the structure of an image recognition device according to an exemplary embodiment of the present disclosure;

[0031] Fig.11 It is a schematic structural diagram of another material sorting equipment shown as an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] The specific implementation methods of the present disclosure will be described below. It should be noted that in the specific description of these implementation methods, in order to provide a concise description, it is impossible for this specification to provide a detailed description of all the features of the actual implementation methods. It should be understood that in the actual implementation of any implementation method, just as in the process of any engineering project or design project, in order to achieve the specific goals of the developer and to meet system-related or business-related restrictions, various specific decisions are often made, and this will also change from one implementation method to another. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the content disclosed by the present disclosure, some changes in design, manufacturing or production based on the technical content disclosed in the present disclosure are just conventional technical means, and should not be understood as insufficient content of the present disclosure.

[0033] Unless otherwise defined, the technical terms or scientific terms used in the present disclosure shall have the usual meanings understood by persons with ordinary skills in the technical field to which the present disclosure belongs. The words "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. Words such as "one" or "one" and the like do not indicate a quantitative limitation, but rather indicate the presence of at least one. Words such as "include" or "comprise" and the like mean that the elements or objects appearing before "include" or "comprise" cover the elements or objects listed after "include" or "comprise" and their equivalents, and do not exclude other elements or objects. Words such as "connect" or "connected" and the like are not limited to physical or mechanical connections, nor are they limited to direct or indirect connections.

[0034] When the material sorting equipment is transmitting materials, in order to ensure the stability of material transmission and prevent the materials from rolling during the transmission process, some evenly distributed convex points are arranged on the conveyor belt to control the relative movement state of the materials on the conveyor belt through multiple convex points.

[0035] In the process of material transmission, the currently transmitted material can generally be determined by performing image recognition processing on the collected images. However, in the actual transmission process, part of the material may adhere to the convex points, and then in the process of image recognition, the convex points on the conveyor belt may be recognized as part of the material, causing interference, which in turn affects the accuracy of material recognition and affects the material recognition results.

[0036] For example, the material sorting device is a plastic sorting device. The structure of the plastic sorting device 100 can be as follows: Figure 1As shown, it includes a feeding mechanism 110, a transmission mechanism 120, a detection mechanism 130, a sorting device 140 and a collecting device 150. The feeding mechanism 110 is used to feed the plastic to be sorted into the transmission mechanism 120. The transmission mechanism 120 at least includes a conveyor belt for conveying the plastic fed by the feeding mechanism 110, and a detection mechanism 130 for detecting the plastic conveyed on the transmission mechanism 120 to identify the plastic currently being conveyed. Among them, the detection mechanism 130 can be a device that uses near-infrared plastic recognition technology to detect the plastic conveyed on the transmission mechanism 120. For example, an infrared camera is used to collect an image to be processed of the plastic currently being conveyed by the conveyor belt, and then the image recognition processing is performed on the image to be processed to determine the contour of the plastic currently being conveyed by the conveyor belt, and then the plastic currently being conveyed is identified based on the obtained contour result. The sorting device 140 is used to remove impurities in the plastic. The collecting device 150 is used to collect the plastic that needs to be retained and the impurities in the plastic.

[0037] During the plastic transmission process, the plastic can be transmitted in the form of particles or sheets. If the plastic is small in size or too thin, and it adheres to the bumps on the conveyor belt during the transmission process, the bumps are easily misidentified as part of the plastic during the image recognition process, thereby affecting the accuracy of plastic recognition and the plastic recognition results. The situation of adhesion may include but is not limited to: plastic particles are arranged vertically above the bumps, or plastic sheets partially cover the bumps on the conveyor belt, etc.

[0038] In the related art, in order to improve the accuracy of convex point recognition, a real-time matching method can be used to determine the convex point image position information corresponding to the image to be processed by template matching, and then eliminate or reduce the interference of the convex point image position on the contour information of the material by information matching, so as to achieve the purpose of improving the accuracy of material recognition. However, using this method to determine the convex point position will affect the material recognition efficiency, thereby affecting the material sorting performance of the material sorting equipment.

[0039] To solve the above problems, an exemplary embodiment of the present disclosure provides an image recognition method applied to a material sorting device. The material sorting device includes a conveyor belt with multiple rows of convex points, and the conveyor belt is used to convey materials. Figure 2 As shown, the image recognition method may include the following steps S210 to S240:

[0040] Step S210, acquiring an image to be processed when the material is being transported by a conveyor belt.

[0041] The image to be processed refers to the image collected during the material sorting equipment transfers the material through the conveyor belt. By obtaining the image to be processed, the current material transfer status can be clarified, and then the currently transferred material can be sorted in a targeted manner.

[0042] The conveyor belt includes multiple rows of protrusions to ensure that the relative motion state of the material on the conveyor belt can be controlled by the protrusions on the conveyor mechanism during the conveying process. In some examples, the number of columns of protrusions in each row is the same and corresponds one to one.

[0043] Step S220, determining a first salient point image in the image to be processed, and determining first contour information of the material.

[0044] In order to determine the image content of the image to be processed, image recognition processing is performed on the image to be processed, and then the first convex point image and the first contour information of the material are obtained. Among them, the first convex point image can be understood as the image corresponding to the convex points of the conveyor belt that are not covered by the material in the acquisition area when the image to be processed is collected. The first contour information can be understood as information obtained by preliminary identification of the location of the material. The first contour information can be obtained by processing the image to be processed with a contour extraction algorithm, for example, a semantic segmentation algorithm, an edge detection algorithm, a connected area marking algorithm, etc. The specific contour extraction algorithm used can be determined according to actual needs. In some examples, the algorithm used to determine the first convex point image can be the same as or different from the algorithm used to determine the first contour information of the material. It can be determined according to actual needs, and the order of determining the first convex point image and the first contour information is not limited in the present disclosure.

[0045] Step S230, based on the matching result between the first salient point image and the reference salient point image, determine the target salient point position information corresponding to the image to be processed.

[0046] The reference convex point image is a convex point splicing image of a plurality of rows of convex points on the conveyor belt. That is, the reference convex point image is a pre-created splicing image. By referring to the convex point image, the position information of the convex point images of the plurality of rows of convex points relative to each other can be clearly determined within one operation cycle of the conveyor belt.

[0047] The first salient point image can be understood as a salient point image that has been correctly identified. However, in the actual image acquisition process, there may be salient points that are partially covered by the material in the acquisition area, and the salient point images corresponding to such salient points may affect the material recognition results. Since the position information of each salient point image in the reference salient point image is known, in order to determine the position information of the salient point images corresponding to each salient point in the acquisition area in the image to be processed, the first salient point image is matched with the reference salient point image, and the distribution of the salient point images corresponding to the multiple salient points in the acquisition area in the image to be processed is quickly determined based on the matching results, thereby obtaining the target salient point position information corresponding to the image to be processed.

[0048] Step S240, based on the matching result between the first contour information and the target salient point position information, the first contour information is updated to obtain the target contour information of the material, so as to identify the material through the target contour information.

[0049] Since the convex points are fixed on the conveyor belt, and during the process of conveying the material, the material may be adhered to the convex points, and when the contour of the image to be processed is extracted, the convex point image may be regarded as part of the material, which may cause the first contour information to deviate from the actual contour information of the material.

[0050] Therefore, in order to improve the accuracy of material identification, the first contour information is matched with the target convex point position information to determine whether there is a situation where the contour is misidentified due to the presence of the convex point image based on the obtained matching result. After obtaining the matching result, updating the first contour information based on the matching result can make the updating process more targeted, effectively eliminate the influence of the convex point image on material identification, and thus improve the accuracy and reliability of contour extraction, so that the obtained target contour information can be more consistent with the real contour of the material, so that when the target contour information is used to identify the material in the future, the recognition efficiency can be improved, and the material recognition performance of the material sorting equipment can be guaranteed.

[0051] According to the material identification method provided by the present invention, by obtaining in advance a reference convex point image corresponding to the multiple rows of convex points included in the conveyor belt, the distribution of each convex point image can be clarified, and then after obtaining the first convex point image of the image to be processed, the target convex point position information corresponding to the image to be processed can be quickly determined through the matching result between the first convex point image and the reference convex point image, thereby improving the efficiency of information determination, and then combined with the matching result with the first contour information of the material, the interference of the convex point image on the first contour information can be quickly eliminated or reduced, and the recognition quality of the target contour information can be improved, thereby helping to promote the material identification process, improve the material sorting efficiency, and ensure the material identification performance of the material sorting equipment.

[0052] In some embodiments, if the image to be processed is the first frame image, then Figure 3 As shown, the above step S230 may include the following steps:

[0053] Step S231, based on the matching result between the first salient point image and the reference salient point image, determine the first row interval salient point image corresponding to the image to be processed in the reference salient point image.

[0054] In order to improve the efficiency of determining the position information and narrow the screening range, the first salient point image is matched with the reference salient point image, and the area image in the reference salient point image that is closest to the first salient point image is determined to obtain a matching result.

[0055] Since the convex points on the conveyor belt are arranged in rows and the arrangement has a certain regularity, by matching the first convex point image with the reference convex point image, it is possible to determine which rows of convex point images in the reference convex point image are closest to the first convex point image, and thus use the row of convex point images closest to the first convex point image as the first row interval convex point image, so that the distribution of multiple convex point images in the image to be processed can be clarified through the first row interval convex point image.

[0056] Step S232, determining first reference position information of each salient point image in the first row interval salient point image according to the reference position information of each salient point image in the reference salient point image.

[0057] The position information of each salient point image in the reference salient point image is relatively fixed relative to each other. Therefore, by referring to the salient point image, the position distribution of salient point images of multiple rows of salient points in the reference salient point image can be determined, thereby determining the reference position information of each salient point image in the reference salient point image.

[0058] When the reference position information of each salient point image in the reference salient point image is predetermined, the first reference position information of each salient point image in the first row interval salient point image can be quickly located and determined according to the determined first row interval salient point image.

[0059] Step S233, based on the first reference position information of each salient point image in the first row interval salient point image, determine the target salient point position information corresponding to the image to be processed.

[0060] The distribution of each convex point image in the target row convex point image can characterize the distribution of convex points on the conveyor belt when the image to be processed is acquired, and then the target convex point position information corresponding to the image to be processed can be obtained by clarifying the first reference position information of each convex point image in the first row interval convex point image. The target convex point position information includes the position information of the first convex point image identified in the image to be processed, and the position information of the convex point image not identified in the image to be processed. The position information of the convex point image not identified in the image to be processed can be understood as the image position information corresponding to the convex points that are completely or partially covered by the material in the acquisition area during image acquisition.

[0061] In other embodiments, if the image to be processed is not the first frame image, then Figure 4 As shown, the above step S230 may include the following steps:

[0062] Step S234, obtaining the position of the second row interval salient point image corresponding to the previous frame image and the reference salient point image in the reference salient point image.

[0063] Since the image to be processed is not the first frame image, and the distribution positions of each row of salient point images are relatively fixed within one operating cycle of the conveyor belt, in order to improve the efficiency of determining the reference position information, the position of the second row of salient point images corresponding to the previous frame image and the image to be processed in the reference salient point image is obtained.

[0064] Step S235, based on the position of the second row interval salient point image in the reference salient point image and the image acquisition order of the image to be processed, determine the third row interval salient point image in the reference salient point image corresponding to the image to be processed.

[0065] When the image acquisition size of the frame images is the same, the number of salient image rows contained in each frame image may be different within one operation cycle of the conveyor belt. Since the previous frame image is adjacent to the image to be processed and the acquisition time is earlier than that of the image to be processed. Therefore, the number of salient image rows corresponding to the image to be processed can be determined through the second row interval salient image, and then according to the position of the second row interval salient image in the reference salient image and the image acquisition order of the image to be processed, the row interval salient image corresponding to the image to be processed is obtained downward along the position of the second row interval salient image in the reference salient image, thereby obtaining the third row interval salient image corresponding to the image to be processed in the reference salient image. For example, the reference salient image includes 128 rows of pixels, and the position of the second row of salient image in the reference salient image is the 50th row of pixels, and the height of the second row interval salient image is 50 rows of pixels. Then, according to the acquisition order of the images to be processed, it can be determined that the position of the image to be processed in the reference salient image is the 100th row, and then according to the position of the image to be processed in the reference salient image, it can be determined that the 101st to 128th rows of pixels in the reference salient image are the third row interval salient image corresponding to the image to be processed, and the 1st to 22nd rows in the next frame of the reference salient image also belong to the row interval salient image corresponding to the image to be processed.

[0066] By determining the salient point image of the third row interval in this way, the difficulty of image positioning can be simplified, the efficiency of determining position information can be improved, and the processing efficiency of material identification can be accelerated.

[0067] Step S236, determining second reference position information of each salient point image in the salient point image of the third row interval according to the reference position information of each salient point image in the reference salient point image.

[0068] When the reference position information of each salient point image in the reference salient point image is predetermined, the second reference position information of each salient point image in the third row interval salient point image can be quickly located and determined according to the determined third row interval salient point image.

[0069] Step S237, determining the target salient point position information corresponding to the image to be processed based on the second reference position information of each salient point image in the third row interval salient point images.

[0070] The distribution of each convex point image in the target row convex point image can characterize the distribution of convex points on the conveyor belt when the image to be processed is acquired, and then the target convex point position information corresponding to the image to be processed can be obtained by clarifying the second reference position information of each convex point image in the third row interval convex point image. The target convex point position information includes the position information of the first convex point image identified in the image to be processed, and the position information of the convex point image not identified in the image to be processed. The position information of the convex point image not identified in the image to be processed can be understood as the image position information corresponding to the convex points that are completely or partially covered by the material in the acquisition area during image acquisition.

[0071] The following embodiments will illustrate the specific determination process of the reference salient point image. Figure 5 As shown, the method for determining the reference salient point image includes:

[0072] Step S310, acquiring a plurality of continuous first image frames when the conveyor belt is running in a no-load state.

[0073] Due to the limited size of each frame of image, each frame of image can only capture images of some rows of convex points when the conveyor belt is running in the unloaded state, and the number of rows of convex points when the conveyor belt is running in one operation cycle cannot be determined. Therefore, in order to ensure the integrity and accuracy of data collection, multiple consecutive frames of the first image when the conveyor belt is running in the unloaded state are obtained to determine the number of rows of convex points transmitted by the conveyor belt in one operation cycle and the distribution of each convex point.

[0074] Step S320, stitching multiple frames of first images to obtain an intermediate image.

[0075] Since the conveyor belt rotates in circles in the material sorting machine, in order to determine the image height corresponding to one rotation of the belt, the acquired multiple frames of images are spliced ​​to obtain an intermediate image, so as to observe the situation on the conveyor belt more comprehensively. Among them, the number of convex dot image rows in the intermediate image is greater than the number of convex dot rows of the conveyor belt to ensure the integrity and accuracy of the reference convex dot image. One rotation of the belt is equivalent to one operation cycle.

[0076] Step S330, extracting a reference salient point image corresponding to the multiple rows of salient points from the intermediate image.

[0077] Since the number of convex point image rows in the intermediate image is greater than the number of convex point rows when the conveyor belt runs in one operating cycle, in order to improve the efficiency of determining the convex point positions in the future, a reference convex point image corresponding to multiple rows of convex points is extracted from the intermediate image, so that the distribution of the convex point images in each image to be processed can be quickly determined through the reference convex point image, thereby improving the efficiency of material recognition.

[0078] According to the method for determining the reference salient point image provided by the present disclosure, the acquisition integrity and reliability of the reference salient point image can be guaranteed.

[0079] In some embodiments, Figure 6 As shown, the above step S330 may include the following steps:

[0080] Step S331, matching each row of salient point images in the intermediate image with a plurality of salient point detection templates respectively to obtain a similarity vector of each row of salient point images.

[0081] Among them, multiple salient point detection templates are determined based on the background image corresponding to multiple rows of salient points, and the salient point detection template corresponds to each column of salient point images. That is, in the process of actual image acquisition, the image acquisition device may be affected by factors such as the environment, light source or its own hardware, resulting in uneven distribution of pixel brightness in the image to be processed. Therefore, the image acquisition device can be used to collect and create a background image in advance when the conveyor belt is running without any object. Since the salient points are evenly distributed on the conveyor belt, it can be clearly observed on the background image that each salient point image is evenly distributed in each row and each column. In order to improve the detection accuracy and reduce the influence of factors such as the environment, light source or its own hardware, a corresponding salient point detection template is created for each image column corresponding to each column of salient point images, so that the detection accuracy of the salient point image can be improved during subsequent detection. The image column may include at least one pixel column. Each image column corresponds to a salient point detection template, and then the salient point detection template can be used to perform targeted detection on all salient point images in its corresponding image column. In one example, the salient point images corresponding to the multiple salient point detection templates may be multiple salient point images in the same row in the background image, thereby simplifying the creation process of the salient point detection templates and improving the creation efficiency. In another example, the salient point images corresponding to the multiple salient point detection templates may not be multiple salient point images in the same row in the background image, but may be the salient point images in the corresponding image column that best represent the image features of the image column, thereby quickly determining each salient point image in the corresponding image column when performing template matching later.

[0082] In order to determine the distribution of each salient point image in the intermediate image, each row of salient point images in the intermediate image is matched with multiple salient point detection templates to obtain the similarity between each row of salient points and its corresponding salient point detection template, thereby obtaining a similarity vector of each row of salient point images. That is, for the same row of salient point images, each column of salient point images is matched with its corresponding salient point detection template to obtain the similarity value between each column of salient point images and the corresponding salient point detection template, and then according to the arrangement order of each salient point image in the same row of salient point images, a similarity vector that can characterize the row of salient point images and multiple salient point detection templates can be obtained.

[0083] Step S332, based on the similarity vector of each row of salient point images in the intermediate image, determine the reference salient point images corresponding to the multiple rows of salient points in the intermediate image.

[0084] Since the similarity vectors of row convex point images collected for different rows of convex points are different in the same operation cycle, the similarity vectors of row convex point images collected for the same row of convex points are approximately the same. Therefore, based on the similarity vector of each row of convex point images in the intermediate image, it is possible to identify whether there is a row convex point image collected for the same row of images in the intermediate image. If different rows of convex point images with the same similarity vector appear in the intermediate image, it can be considered that the conveyor belt has completed an operation cycle, and then the multiple rows of convex points corresponding to one rotation of the conveyor belt can be determined, thereby obtaining a reference convex point image that can characterize the distribution of multiple rows of convex points when the conveyor belt rotates one circle.

[0085] In some examples, the above step S332 may include the following steps:

[0086] Step a1, determining the target row of salient point images having the greatest similarity to the first row of salient point images among the other rows of salient point images in the intermediate image.

[0087] Step a2, determining reference salient point images corresponding to the multiple rows of salient points in the intermediate image according to the positions of the target row salient point images in the intermediate image.

[0088] Specifically, to ensure the integrity of the reference image, the similarity vectors of the first row of salient point images in the intermediate image are matched with the similarity vectors of other rows of salient point images in the intermediate image, and the similarities between the first row of salient point images and other rows of salient point images are determined, and the row of salient point images with the greatest similarity to the first row of salient point images in other rows of salient point images are used as target row of salient point images. For example, by comparing the Euclidean distance between the similarity vectors of the first row of salient point images and the similarity vectors of other rows of salient point images in the intermediate image, the row of salient point images closest to the first row of salient point images can be identified, and the row of salient point images can be used as the target row of salient point images that best matches the first row of salient point images.

[0089] Since the intermediate image contains more than one row of convex points in an operation cycle, it can be considered that when the similarity between the target row convex point image and the first row convex point image is the largest, the convex point row corresponding to the target row convex point image and the convex point row of the first row convex point image are considered to be the same row. Therefore, according to the position of the target row convex point image in the intermediate image, the reference convex point image corresponding to the multiple rows of convex points in the intermediate image is determined. The reference convex point image includes the convex point image of the previous row from the first row convex point image to the target row convex point image. For example, the first row convex point image is the 0th row convex point image in the intermediate image, and the target row convex point image is the 100th row convex point image in the intermediate image, then the reference convex point image includes the convex point images of the 0th row to the 99th row.

[0090] Step S333, extracting a reference salient point image.

[0091] In some embodiments, Figure 7 As shown, the above step S330 may also include the following steps:

[0092] Step S334, obtaining a plurality of consecutive second images when the conveyor belt is running in an unloaded state, wherein the plurality of second images can be obtained in the same manner as the plurality of first images, which will not be described in detail here.

[0093] Step S335, sequentially matching each frame of the second image with the intermediate image, determining an intermediate local image in the intermediate image corresponding to the second image, and determining a reference salient point image corresponding to the multiple rows of salient points based on a height relationship between two adjacent intermediate local images.

[0094] Since the second image and the first image are acquired in the same manner, each frame of the second image is sequentially matched with the previously spliced ​​intermediate image, and an intermediate partial image that can match the second image can be determined from the intermediate image. In the same operating cycle, the height relationship between two adjacent intermediate partial images can be the same, so the reference salient point image corresponding to multiple rows of salient points in the same operating cycle can be determined based on the height relationship between the two adjacent intermediate partial images.

[0095] In some examples, the height of each frame of the second image is the same, and the above step S335 may include the following steps:

[0096] Step b1, matching the current second image with the intermediate image, determining a first intermediate partial image in the intermediate image corresponding to the current second image, and first position information of the first intermediate partial image, wherein the first position information includes a first height of the first intermediate partial image;

[0097] Step b2: determining second position information of the second intermediate partial image.

[0098] Step b3, if the difference between the second height and the first height is equal to the frame image height, performing a step of matching the next second image with the intermediate image;

[0099] Step b4: If the difference between the second height and the first height is less than the frame image height, a reference salient point image corresponding to multiple rows of salient points in the intermediate image is determined based on the acquisition order of the current second image in multiple frames of second images, the first height and the second height.

[0100] Specifically, feature matching is performed between the current second image and the intermediate image to find the corresponding part in the image. After the matching is successful, the corresponding area of ​​the current second image in the intermediate image can be determined, and this area is called the first intermediate partial image. The position information of the first intermediate partial image is recorded, including its first height, that is, the distance from the bottom of the intermediate image to the bottom of the first intermediate partial image.

[0101] To determine the height relationship between two adjacent intermediate partial images, second position information of the second intermediate partial image is determined, wherein the second intermediate partial image is a partial image in the intermediate image corresponding to the previous second image, and the second position information includes a second height of the second intermediate partial image.

[0102] If the difference between the second height and the first height is equal to the frame image height, it indicates that the current second image and the previous second image belong to the same operating cycle, so the second image of the next frame can be processed.

[0103] If the difference between the second height and the first height is less than the frame image height, it is characterized that the current second image includes two operation cycles. Therefore, based on the acquisition order of the current second image in multiple frames of second images, the first height and the second height, the reference salient point image corresponding to the multiple rows of salient points in the intermediate image is determined. Specifically, according to the first height and the second height, the target image height in the intermediate image that belongs to the same operation cycle as the previous second image can be determined, and then according to the acquisition order of the current second image in multiple frames of second images, the image height of the second image and the target image height, it is acquired from the top of the intermediate image downward to obtain a reference salient point image. For example, the target image height = the first height + the image height of the second image - the second height.

[0104] Step S336, extracting a reference salient point image from the intermediate image.

[0105] The following embodiment will illustrate the specific process of determining the reference position information of each salient point image in the reference salient point image:

[0106] Step c1, determining reference position information of the current salient point image in the reference salient point image;

[0107] If the current salient point image is not a first row salient point image, determining the reference position information of the current salient point image according to the first row spacing between the row where the current salient point image is located and the previous row;

[0108] Step c2: if the current salient point image is the first row of salient point images, determine the second row spacing between the last row of salient point images of the reference salient point image and the next first row of salient point images through the intermediate image, and determine the reference position information of the current salient point image according to the second row spacing.

[0109] Specifically, when the reference position information is clear, in order to facilitate the subsequent material identification process, the position information of each salient point image in the image to be processed can be quickly located, and the position information of each salient point image in the reference image is determined respectively.

[0110] For example, starting from the second row of salient point images, the first row spacing between the row where the current salient point image is located and the previous row can be determined respectively, and then the reference position information of the current salient point image can be determined according to the first row spacing.

[0111] Since the conveyor belt is always rotating, the last row of convex dot images in the previous operation cycle is the previous row of convex dot images in the current operation cycle. Therefore, in order to accurately calculate the subsequent convex dot positions, the second row spacing between the last row of convex dot images of the reference convex dot image and the next first row of convex dot images is determined through the intermediate image, and the reference position information of the current convex dot image is determined based on the second row spacing to ensure the continuity and stability of the recognition process.

[0112] In some examples, if the reference salient image is determined based on a similarity matching method of the salient image, the storage expression for the position information of each salient image in the reference salient image may be: Start=0; End=min(blocks_loc[p][j])+min(d[1][j])-min(blocks_loc[1][j]); wherein d[1][j] represents the distance stored in the 1st row and the jth column; blocks_loc[p][j] represents the upper left corner coordinate of the salient point stored in the pth row and the jth column.

[0113] In other examples, if the reference salient image is determined based on image feature matching, the storage expression of the reference salient image may be: d[p][j]=image height-blocks_loc[p][j]+blocks_loc[1][j].

[0114] In some embodiments, the above step S240 may include: if the target salient point position partially overlaps with the first contour information, then the contour information in the first contour information that corresponds to the target salient point position and overlaps with it is removed to obtain the target contour information of the material. Through the matching results of the first contour information and multiple target salient point positions, it can be determined whether there is a situation in which the salient point image is mistakenly regarded as part of the material during the process of identifying the material contour. If it is determined based on the matching results that there is a target salient point position that partially overlaps with the first contour information, then the contour information in the first contour information that corresponds to the target salient point position and overlaps with it is removed to reduce image noise, thereby eliminating the influence of the salient point image on material identification, and then obtaining target contour information that can better fit the actual contour of the material, thereby improving the accuracy of the target contour information. For example, the first contour information of the material obtained is as follows: Figure 8 When it is determined that there is a target salient point position partially overlapping with the first contour information, the contour information corresponding to the target salient point position in the first contour information is removed, and the target contour information obtained can be as follows: Fig. 9 shown.

[0115] In other embodiments, if there are multiple first contour information and there is first contour information that partially overlaps with the target salient point position, the first contour information is discarded to avoid the influence of the first contour information on material identification, thereby simplifying the difficulty of material identification, and effectively improving the accuracy of material identification and ensuring the quality of material identification.

[0116] In some optional implementation scenarios, the process of identifying the material (plastic) currently transmitted by the material sorting device through the collected image to be processed can be as follows:

[0117] A reference salient point image is created in advance, and distribution information of each row of salient point images is determined based on the reference salient point image.

[0118] In the actual image recognition process, the belt is first allowed to idle for a short period of time, as long as the idling time is greater than the acquisition time of one frame of image, and then the plastic is placed.

[0119] Determine the first salient point image in the first frame of the image to be processed, and determine the first contour information of the material. The first contour information of the plastic in the image to be processed can be obtained through infrared image detection.

[0120] The first salient point image is matched with the reference salient point image to determine the first row interval salient point image in the reference salient point image corresponding to the first frame of the image to be processed, and then the first reference position information of each salient point image in the first row interval salient point image is determined according to the reference position information of each salient point image in the reference salient point image, so that the target salient point position information corresponding to the first frame of the image to be processed can be determined.

[0121] After obtaining the target salient point position information, the first contour information is matched with the target salient point position information. If there is Figure 8 The target position information partially overlaps with the first contour information as shown, then the contour information corresponding to the overlap with the target position information in the first contour information is removed, and the following is obtained: Fig. 9 Target profile information for the plastics shown.

[0122] Based on the first row of interval salient point images, the second row of interval salient point images corresponding to the next image to be processed in the reference salient point image can be determined, and then the target salient point position information corresponding to the next image to be processed can be quickly determined without the need for secondary re-matching and determination, thereby effectively improving the positioning efficiency of the salient point image, facilitating the subsequent identification and processing of plastics, improving material identification efficiency, and ensuring the performance of material sorting equipment.

[0123] Based on the same inventive concept, the present disclosure also provides an image recognition device applied to a material sorting device. The material sorting device includes a conveyor belt with multiple rows of convex points. Fig.10 As shown, the image recognition device 400 may include: a first acquisition module 410 , a first determination module 420 , a second determination module 430 and an update module 440 .

[0124] A first acquisition module 410 is used to acquire an image to be processed when a material is being transported by a conveyor belt;

[0125] A first determination module 420, used to determine a first salient point image in the image to be processed, and to determine first contour information of a material;

[0126] A second determination module 430 is used to determine target salient point position information corresponding to the image to be processed based on a matching result between the first salient point image and a reference salient point image, wherein the reference salient point image is a salient point splicing image containing multiple rows of salient points on a conveyor belt;

[0127] The updating module 440 is used to update the first contour information based on the matching result between the first contour information and the target salient point position information, and obtain the target contour information of the material, so as to identify the material through the target contour information.

[0128] In some embodiments, if the image to be processed is the first frame image, the second determination module 430 may include: a first determination unit, used to determine the first row interval salient point image corresponding to the image to be processed in the reference salient point image based on the matching result between the first salient point image and the reference salient point image; a second determination unit, used to determine the first reference position information of each salient point image in the first row interval salient point image according to the reference position information of each salient point image in the reference salient point image; a third determination unit, used to determine the target salient point position information corresponding to the image to be processed based on the first reference position information of each salient point image in the first row interval salient point image.

[0129] In some embodiments, if the image to be processed is not the first frame image, the second determination module 430 may include: a position acquisition unit, used to obtain the position of the second row interval salient point image corresponding to the reference salient point image of the previous frame image in the reference salient point image; a fourth determination unit, used to determine the third row interval salient point image corresponding to the image to be processed in the reference salient point image based on the position of the second row interval salient point image in the reference salient point image and the image acquisition order of the image to be processed; a fifth determination unit, used to determine the second reference position information of each salient point image in the third row interval salient point image according to the reference position information of each salient point image in the reference salient point image; and a sixth determination unit, used to determine the target salient point position information corresponding to the image to be processed based on the second reference position information of each salient point image in the third row interval salient point image.

[0130] In some embodiments, the device for determining the reference salient point image includes: a second acquisition module, used to acquire multiple consecutive frames of first images when the conveyor belt is running in a no-load state; a splicing module, used to splice multiple frames of first images to obtain an intermediate image, and the number of salient point image rows in the intermediate image is greater than the number of salient point rows of the conveyor belt; an extraction module, used to extract the reference salient point image corresponding to multiple rows of salient points from the intermediate image.

[0131] In some embodiments, the extraction module includes: a first matching unit, used to match each row of salient point images in the intermediate image with multiple salient point detection templates respectively to obtain a similarity vector of each row of salient point images, wherein the multiple salient point detection templates are determined based on the background image corresponding to the multiple rows of salient points, and the salient point detection templates correspond to each column of salient point images; a first screening unit, used to determine the reference salient point images corresponding to the multiple rows of salient points in the intermediate image based on the similarity vector of each row of salient point images in the intermediate image; and a first extraction unit, used to extract the reference salient point images.

[0132] In some embodiments, the first screening unit includes: a first execution unit, used to determine a target row of salient point images having the greatest similarity with a first row of salient point images among other rows of salient point images in the intermediate image; and a second execution unit, used to determine a reference salient point image corresponding to multiple rows of salient points in the intermediate image according to a position of the target row of salient point images in the intermediate image, the reference salient point images including a row of salient point images from the first row of salient point images to the previous row of salient point images of the target row of salient point images.

[0133] In some embodiments, the extraction module includes: an image acquisition unit, used to acquire multiple consecutive frames of second images when the conveyor belt is running in a no-load state; a second screening unit, used to sequentially match each frame of the second image with the intermediate image, determine the intermediate local image corresponding to the second image in the intermediate image, and determine the reference salient point image corresponding to multiple rows of salient points based on the height relationship between two adjacent intermediate local images; a second extraction unit, used to extract the reference salient point image from the intermediate image.

[0134] In some embodiments, the height of each frame of the second image is the same; the second screening unit includes: a third execution unit, which is used to match the current second image with the intermediate image, determine the first intermediate local image corresponding to the current second image in the intermediate image, and the first position information of the first intermediate local image, the first position information includes the first height of the first intermediate local image; a fourth execution unit, which is used to determine the second position information of the second intermediate local image, the second intermediate local image is the local image in the intermediate image corresponding to the previous second image, and the second position information includes the second height of the second intermediate local image; a fifth execution unit, which is used to perform the step of matching the next second image with the intermediate image if the difference between the second height and the first height is equal to the frame image height; a sixth execution unit, which is used to determine the reference salient point image corresponding to multiple rows of salient points in the intermediate image based on the acquisition order of the current second image in multiple frames of second images, the first height and the second height if the difference between the second height and the first height is less than the frame image height.

[0135] In some embodiments, a device for determining reference position information of each salient point image in a reference salient point image includes: a first information determination module, for determining the reference position information of the current salient point image according to a first line spacing between a line where the current salient point image is located and a previous line if the current salient point image is not a first line salient point image; and a second information determination module, for determining a second line spacing between a last line salient point image of the reference salient point image and a next first line salient point image through an intermediate image if the current salient point image is a first line salient point image, and determining the reference position information of the current salient point image according to the second line spacing.

[0136] In some embodiments, the updating module 440 includes: an information updating unit, which is used to remove the contour information corresponding to the target convex point position in the first contour information if the target convex point position partially overlaps with the first contour information, so as to obtain the target contour information of the material.

[0137] Regarding the image recognition device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0138] Based on the same inventive concept, Fig.11 As shown, one embodiment of the present disclosure provides a material sorting device 500. The material sorting device 500 includes a memory 510, a processor 520, and an input / output (I / O) interface 530. The memory 510 is used to store instructions. The processor 520 is used to call the instructions stored in the memory 510 to execute the image recognition method of the embodiment of the present disclosure. The processor 520 is connected to the memory 510 and the I / O interface 530, respectively, for example, through a bus system and / or other forms of connection mechanisms (not shown). The memory 510 can be used to store programs and data, including the program of the image segmentation method involved in the embodiment of the present disclosure. The processor 520 executes various functional applications and data processing of the material sorting device 500 by running the program stored in the memory 510.

[0139] In the embodiment of the present disclosure, the processor 520 can be implemented in at least one hardware form of a digital signal processor (Digital Signal Processing, DSP), a field programmable gate array (Field-Programmable Gate Array, FPGA), and a programmable logic array (Programmable Logic Array, PLA). The processor 520 can be a central processing unit (Central Processing Unit, CPU) or one or a combination of other forms of processing units with data processing capabilities and / or instruction execution capabilities.

[0140] The memory 510 in the embodiment of the present disclosure may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), flash memory, hard disk drive (HDD) or solid-state drive (SSD).

[0141] In the embodiment of the present disclosure, the I / O interface 530 may be used to receive input instructions (such as digital or character information, and key signal input related to user settings and function control of the material sorting device 500, etc.), and may also output various information to the outside (such as images or sounds, etc.). In the embodiment of the present disclosure, the I / O interface 530 may include one or more of a physical keyboard, a function key (such as a volume control key, a switch key, etc.), a mouse, a joystick, a trackball, a microphone, a speaker, and a touch panel.

[0142] Based on the same inventive concept, the present disclosure further provides a computer-readable storage medium, which stores the following program, and the program is used to execute the image recognition method of any of the aforementioned embodiments.

[0143] The present disclosure uses specific words to describe the embodiments of the present disclosure. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present disclosure. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different locations in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of the present disclosure may be appropriately combined.

[0144] In the context of the present disclosure, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not refer to the singular, but may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.

[0145] Similarly, it should be noted that in order to simplify the description of the disclosure and thus help understand one or more application embodiments, in the above description of the embodiments of the disclosure, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of the disclosure are more than the features required for protection. In fact, the features of the embodiment are less than all the features of the single embodiment disclosed above.

[0146] The basic concepts have been described above. Obviously, for those skilled in the art, the above disclosure is only an example and does not constitute a limitation of the present disclosure. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and corrections to the present disclosure. Such modifications, improvements and corrections are suggested in the present disclosure, so such modifications, improvements and corrections still belong to the spirit and scope of the embodiments of the present disclosure.

Claims

1. An image recognition method, characterized in that: Applied to a material sorting device, the material sorting device comprises a conveyor belt with multiple rows of convex points, the conveyor belt is used to convey materials, and the method comprises: Acquire an image to be processed when the material is transmitted through the conveyor belt; Determine a first salient point image in the image to be processed, and determine first contour information of the material; Determine target salient point position information corresponding to the image to be processed based on a matching result between the first salient point image and a reference salient point image, wherein the reference salient point image is a salient point splicing image of the conveyor belt containing the multiple rows of salient points; Based on the matching result between the first contour information and the target salient point position information, the first contour information is updated to obtain the target contour information of the material, so as to identify the material through the target contour information.

2. The image recognition method according to claim 1, characterized in that: If the image to be processed is a first frame image, determining target salient point position information corresponding to the image to be processed based on a matching result between the first salient point image and a reference salient point image includes: Based on the matching result between the first salient point image and the reference salient point image, determining a first row interval salient point image in the reference salient point image corresponding to the image to be processed; Determining first reference position information of each salient point image in the first row interval salient point image according to reference position information of each salient point image in the reference salient point image; Based on the first reference position information of each salient point image in the first row interval salient point images, target salient point position information corresponding to the image to be processed is determined.

3. The image recognition method according to claim 1, characterized in that: If the image to be processed is not the first frame image, then determining the target salient point position information corresponding to the image to be processed based on the matching result between the first salient point image and the reference salient point image includes: Obtaining the position of the second row interval salient point image corresponding to the previous frame image and the reference salient point image in the reference salient point image; Determine, based on the position of the second row interval salient point image in the reference salient point image and the image acquisition order of the image to be processed, a third row interval salient point image in the reference salient point image corresponding to the image to be processed; Determining second reference position information of each salient point image in the third row interval salient point image according to the reference position information of each salient point image in the reference salient point image; Based on the second reference position information of each salient point image in the third row interval salient point images, target salient point position information corresponding to the image to be processed is determined.

4. The image recognition method according to claim 2 or 3, characterized in that: The process of determining the reference salient point image includes: Acquire a plurality of continuous first images when the conveyor belt is running in a no-load state; Splicing the multiple frames of first images to obtain an intermediate image, wherein the number of convex point image rows in the intermediate image is greater than the number of convex point rows of the transmission belt; A reference salient point image corresponding to the plurality of rows of salient points is extracted from the intermediate image.

5. The image recognition method according to claim 4, characterized in that: The step of extracting a reference salient point image corresponding to the plurality of rows of salient points from the intermediate image comprises: Matching each row of salient point images in the intermediate image with a plurality of salient point detection templates respectively to obtain a similarity vector of each row of salient point images, wherein the plurality of salient point detection templates are determined based on background images corresponding to the plurality of rows of salient points, and the salient point detection templates correspond to each column of salient point images; Determining reference salient point images corresponding to the multiple rows of salient points in the intermediate image based on the similarity vector of each row of salient point images in the intermediate image; The reference salient image is extracted.

6. The image recognition method according to claim 5, characterized in that: The determining, based on the similarity vector of each row of salient point images in the intermediate image, reference salient point images corresponding to the multiple rows of salient points in the intermediate image comprises: Determine a target row of salient point images having the greatest similarity to the first row of salient point images among other rows of salient point images in the intermediate image; According to the position of the target row of salient point images in the intermediate image, a reference salient point image corresponding to the multiple rows of salient points in the intermediate image is determined, and the reference salient point image includes a row of salient point images from the first row of salient point images to the previous row of salient point images of the target row of salient point images.

7. The image recognition method according to claim 4, characterized in that: The step of extracting a reference salient point image corresponding to the plurality of rows of salient points from the intermediate image comprises: Acquire a plurality of continuous second images when the conveyor belt is running in a no-load state; Matching each frame of the second image with the intermediate image in sequence, determining an intermediate partial image in the intermediate image corresponding to the second image, and determining the reference salient point image corresponding to the multiple rows of salient points based on a height relationship between two adjacent intermediate partial images; The reference salient image is extracted from the intermediate image.

8. The image recognition method according to claim 7, characterized in that: The height of the second image in each frame is the same; The step of sequentially matching each frame of the second image with the intermediate image, determining an intermediate partial image in the intermediate image corresponding to the second image, and determining the reference salient point image corresponding to the multiple rows of salient points based on a height relationship between two adjacent intermediate partial images, comprises: Matching the current second image with the intermediate image, determining a first intermediate partial image in the intermediate image corresponding to the current second image, and first position information of the first intermediate partial image, wherein the first position information includes a first height of the first intermediate partial image; Determine second position information of a second intermediate partial image, where the second intermediate partial image is a partial image in the intermediate image corresponding to a previous second image, and the second position information includes a second height of the second intermediate partial image; If the difference between the second height and the first height is equal to the frame image height, performing a step of matching the next second image with the intermediate image; If the difference between the second height and the first height is less than the frame image height, the reference salient point image corresponding to the multiple rows of salient points in the intermediate image is determined based on the acquisition order of the current second image in multiple frames of the second images, the first height and the second height.

9. The image recognition method according to claim 4, characterized in that: The process of determining the reference position information of each salient point image in the reference salient point image includes: If the current salient point image is not a first row salient point image, determining reference position information of the current salient point image according to a first row spacing between the row where the current salient point image is located and a previous row; If the current salient point image is a first row of salient point images, a second row spacing between a last row of salient point images of the reference salient point image and a next first row of salient point images is determined through the intermediate image, and reference position information of the current salient point image is determined according to the second row spacing.

10. The image recognition method according to claim 1, characterized in that: The updating of the first contour information based on the matching result between the first contour information and the target convex point position information to obtain the target contour information of the material includes: If the target convex point position partially overlaps with the first contour information, the contour information corresponding to and overlapping with the target convex point position in the first contour information is removed to obtain the target contour information of the material.

11. An image recognition device, characterized in that: Applied to material sorting equipment, the material sorting equipment includes a conveyor belt with multiple rows of convex points, the conveyor belt is used to transport materials, and the device includes: A first acquisition module, used for acquiring an image to be processed when the material is transmitted through the conveyor belt; A first determination module, used to determine a first salient point image in the image to be processed, and to determine first contour information of the material; A second determination module is used to determine the target salient point position information corresponding to the image to be processed based on the matching result between the first salient point image and a reference salient point image, wherein the reference salient point image is a salient point splicing image of the conveyor belt containing the multiple rows of salient points; An updating module is used to update the first contour information based on the matching result between the first contour information and the target salient point position information, so as to obtain the target contour information of the material and identify the material through the target contour information.

12. A material sorting device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the image recognition method according to any one of claims 1 to 10 by executing the computer instructions.

13. A computer-readable storage medium storing the following program, wherein the program is used to execute the image recognition method according to any one of claims 1 to 10.

Images