Material detection method and system of material receiving machine, readable storage medium and computer

Abstract

The invention provides a material detection method and system for a material receiving machine, a readable storage medium and a computer, and the method comprises the steps: collecting video data of a preset region on the material receiving machine in real time, and carrying out the frame processing of the video data, and obtaining a corresponding video frame sequence; performing sequence detection on the video frame sequence by using a preset deep learning model to obtain corresponding target frame information; constructing a material detection model, extracting a target feature corresponding to the target frame information by using the material detection model, and performing state prediction according to the target feature so as to calculate a matching degree between the target feature and preset target data; determining a target type of the target feature according to the matching degree, and obtaining a corresponding detection condition based on the target type; and judging whether the target feature meets the detection condition, and if the target feature meets the detection condition, outputting a judgment result of the target feature.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a material detection method, system, readable storage medium and computer for a material receiving machine. Background Art

[0002] With the rapid development of science and technology and the improvement of people's living standards, the number of manufacturing enterprises is gradually increasing.

[0003] The material receiving machine is one of the commonly used equipment in industrial equipment. It can improve production efficiency, reduce production costs and improve product quality. At present, there are usually only two ways to detect materials in the material receiving machine. One way is for the staff to perform manual screening before the materials enter the material receiving machine. The other way is to use image acquisition equipment to pre-collect pictures of several abnormal materials, use the abnormal pictures to build a corresponding abnormal database, and compare the current materials on the material receiving machine with the abnormal database to realize material detection. However, the manual method often has large errors and needs to rely on the experience of the staff. The method of using the abnormal database is only suitable for relatively simple material detection, which results in the detection accuracy and detection speed not meeting user requirements. Summary of the Invention

[0004] The embodiments of the present application provide a material detection method, system, readable storage medium and computer for a material receiving machine to at least address the deficiencies in the above-mentioned related technologies.

[0005] In a first aspect, an embodiment of the present application provides a material detection method for a material receiving machine, comprising the following steps: Step 1: real-time acquisition of video data of a preset area on the material receiving machine, and frame-processing of the video data to obtain a corresponding video frame sequence; Step 2: Using a preset deep learning model to perform sequence detection on the video frame sequence to obtain corresponding target border information; Step 3: Construct a material detection model, and use the material detection model to extract the target features corresponding to the target frame information, and perform state prediction based on the target features to calculate the matching degree between the target features and the preset target data; Step 4: determining the target type of the target feature according to the matching degree, and obtaining corresponding detection conditions based on the target type; Step 5: Determine whether the target feature meets the detection condition. If the target feature meets the detection condition, output the determination result of the target feature.

[0006] Furthermore, the step 1 includes: Collecting video data of a preset area on the material receiving machine in real time, and processing the video data using a color conversion algorithm to obtain corresponding three-channel image data; The mask minimum circumscribed rectangle of the three-channel image data is calculated based on a preset mask algorithm, and frame alignment processing is performed using a temporal frame processing algorithm to obtain a corresponding video frame sequence.

[0007] Furthermore, the step 2 includes: Constructing a lightweight backbone network and replacing the backbone network of the YOLOv5 model with the lightweight backbone network to obtain an optimized YOLOv5 model; The optimized YOLOv5 model is input into a depthwise separable convolution algorithm to obtain a corresponding deep learning model, and the deep learning model is used to perform sequence detection on the video frame sequence to obtain corresponding target border information.

[0008] Furthermore, the step three includes: Processing the target border information using the material detection model to obtain a Mahalanobis distance and a cosine distance corresponding to the target border information; A weighted combination is performed according to the Mahalanobis distance and the cosine distance, and a matching degree corresponding to the target feature is calculated based on the weighted combination result.

[0009] Furthermore, the step 4 includes: Calculating the matching degree using the Hungarian algorithm to construct a corresponding cost matrix, and constraining the cost matrix to obtain optimal allocation data corresponding to the matching degree; The cost matrix is ​​adjusted according to the optimal allocation data, and the adjusted cost matrix is ​​calculated to determine the target type of the target feature, and the corresponding detection condition is obtained from a preset condition database according to the target type.

[0010] In a second aspect, the present invention further provides a material detection system for a material receiving machine, characterized in that it includes: A data processing module is used to collect video data of a preset area on the material receiving machine in real time, and process the video data frame by frame to obtain a corresponding video frame sequence; A sequence detection module is used to perform sequence detection on the video frame sequence using a preset deep learning model to obtain corresponding target frame information; A state prediction module is used to build a material detection model, and use the material detection model to extract the target features corresponding to the target frame information, and perform state prediction based on the target features to calculate the matching degree between the target features and the preset target data; a condition acquisition module, configured to determine a target type of the target feature according to the matching degree, and acquire corresponding detection conditions based on the target type; The condition judgment module is used to judge whether the target feature meets the detection condition, and if the target feature meets the detection condition, output the judgment result of the target feature.

[0011] Furthermore, the data processing module is specifically used to: Collecting video data of a preset area on the material receiving machine in real time, and processing the video data using a color conversion algorithm to obtain corresponding three-channel image data; The mask minimum circumscribed rectangle of the three-channel image data is calculated based on a preset mask algorithm, and frame alignment processing is performed using a temporal frame processing algorithm to obtain a corresponding video frame sequence.

[0012] Furthermore, the sequence detection module is specifically used to: Constructing a lightweight backbone network and replacing the backbone network of the YOLOv5 model with the lightweight backbone network to obtain an optimized YOLOv5 model; The optimized YOLOv5 model is input into a depthwise separable convolution algorithm to obtain a corresponding deep learning model, and the deep learning model is used to perform sequence detection on the video frame sequence to obtain corresponding target border information.

[0013] Furthermore, the state prediction module is specifically used to: Processing the target border information using the material detection model to obtain a Mahalanobis distance and a cosine distance corresponding to the target border information; A weighted combination is performed according to the Mahalanobis distance and the cosine distance, and a matching degree corresponding to the target feature is calculated based on the weighted combination result.

[0014] Furthermore, the condition acquisition module is specifically used to: Calculating the matching degree using the Hungarian algorithm to construct a corresponding cost matrix, and constraining the cost matrix to obtain optimal allocation data corresponding to the matching degree; The cost matrix is ​​adjusted according to the optimal allocation data, and the adjusted cost matrix is ​​calculated to determine the target type of the target feature, and the corresponding detection condition is obtained from a preset condition database according to the target type.

[0015] In a third aspect, the present invention further proposes a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the material detection method of the material receiving machine as described above.

[0016] In a fourth aspect, the present invention further proposes a computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the material detection method of the material receiving machine as described above when executing the computer program.

[0017] Compared with the related art, the embodiments of the present application provide a material detection method, system, readable storage medium and computer for a material receiving machine. Video is captured from a preset area of ​​the material receiving machine, and the video data is processed frame by frame to obtain a corresponding video frame sequence, and a frame-by-frame analysis method is used to improve the overall material detection accuracy; a deep learning model is used to perform sequence detection on the video frame sequence, and a material detection model is used to perform feature extraction on the target border information obtained by the sequence detection, and a state prediction is performed based on the target feature to calculate the matching degree between the target feature and the preset target data, and the matching degree is used to realize rapid detection of materials, thereby improving the overall detection efficiency, and the matching degree is used to determine the corresponding detection conditions, and the efficient detection of materials is realized through the detection conditions, thereby improving the accuracy of the overall material detection.

[0018] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 Flowchart of the material detection method of the material receiving machine in the first embodiment of the present invention; Figure 2 2 is a structural block diagram of a material detection system of a material receiving machine in a second embodiment of the present invention; Figure 3 FIG. 4 is a structural block diagram of a computer in a third embodiment of the present invention.

[0020] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making any creative efforts are within the scope of protection of this application.

[0022] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0023] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0024] Unless otherwise defined, the technical or scientific terms involved in this application should have the usual meaning understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "the" and the like involved in this application do not indicate a quantitative limitation and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Example 1

[0025] See also Figure 1 , which shows a material detection method for a material receiving machine in the first embodiment of the present invention, and the method specifically includes steps S101 to S104: S101, collecting video data of a preset area on the material receiving machine in real time, and processing the video data frame by frame to obtain a corresponding video frame sequence; Furthermore, the step S101 specifically includes steps S1011 to S1012: S1011, collecting video data of a preset area on the material receiving machine in real time, and processing the video data using a color conversion algorithm to obtain corresponding three-channel image data; S1012 , calculating a minimum circumscribed rectangle of a mask on the three-channel image data based on a preset mask algorithm, and performing frame alignment processing using a temporal frame processing algorithm to obtain a corresponding video frame sequence.

[0026] In a specific implementation, an industrial camera with a resolution of 1280×720 or higher is selected to collect data from a preset area of ​​the feeder, wherein the frame rate is selected to be greater than 60 fps. A ring-shaped LED light source is configured on the industrial camera, projecting at a 45° angle to eliminate reflections. A color conversion algorithm (in this embodiment, the color conversion algorithm includes OpenCV's Bayer format conversion algorithm, a bilinear difference algorithm, a nearest neighbor difference algorithm, etc.) is used to perform spatial conversion on the video data to output a standard BGR three-channel image. Furthermore, a pre-calibrated binary mask is constructed to extract the target area to calculate the minimum bounding rectangle of the mask of the three-channel image data, and a time-series frame processing algorithm is used to process the mask. The minimum circumscribed rectangle of the mask is calculated for the three-channel image data based on a preset mask algorithm, and frame alignment is performed using a temporal frame processing algorithm (in this embodiment, the frame alignment steps include using the Shi-Tomasi corner detection algorithm to detect 200 feature points, and using the Lucas-Kanade pyramid optical flow method to track the displacement of the feature points, using the RANSAC algorithm to estimate the affine transformation matrix, and using the bilinear interpolation algorithm to complete the corresponding frame alignment) to obtain a corresponding video frame sequence.

[0027] S102, performing sequence detection on the video frame sequence using a preset deep learning model to obtain corresponding target border information; Furthermore, the step S102 specifically includes steps S1021 and S1022: S1021, constructing a lightweight backbone network, and replacing the backbone network of the YOLOv5 model with the lightweight backbone network to obtain an optimized YOLOv5 model; S1022: Input the optimized YOLOv5 model into a depthwise separable convolution algorithm to obtain a corresponding deep learning model, and use the deep learning model to perform sequence detection on the video frame sequence to obtain corresponding target bounding box information.

[0028] In the specific implementation, a lightweight backbone network is constructed (in this implementation, any one of ShuffleNetv2, MobileNetV2, GhostNet, or MnasNet is selected as the lightweight backbone network) and replaced with the CSPDarknet53 backbone network of the YOLOv5 model to achieve channel rearrangement and obtain the optimized YOLOv5 model. Furthermore, the convolution algorithm in the standard convolution layer of the optimized YOLOv5 model is replaced with a depthwise separable convolution algorithm. At the same time, a channel attention mechanism is introduced, an SE module is added to the key feature layer, and the Hardswish activation function is used instead of SiLU to reduce the computational complexity to obtain the corresponding deep learning model. The deep learning model is used to perform sequence detection on the video frame sequence obtained above, and non-maximum suppression is performed on the video frame sequence according to the current frame sequence. The border information is extracted based on the result to obtain the corresponding target border information.

[0029] S103, constructing a material detection model, and using the material detection model to extract target features corresponding to the target frame information, and performing state prediction based on the target features to calculate the matching degree between the target features and preset target data; Furthermore, the step S103 specifically includes steps S1031 and S1032: S1031, using the material detection model to process the target border information to obtain the Mahalanobis distance and cosine distance corresponding to the target border information; S1032: Perform weighted combination according to the Mahalanobis distance and the cosine distance, and calculate the matching degree corresponding to the target feature based on the weighted combination result.

[0030] In the specific implementation, a material detection model is constructed, and the target border information is calculated using the material detection model to calculate the Mahalanobis distance corresponding to the target border information: ; Where, Indicates the The preset target data and The Mahalanobis distance between target bounding box information, Indicates the The coordinates corresponding to the target border information, Indicates the The coordinates of the preset target data, Represents the covariance matrix between the preset target data and the target bounding box information; The target border information is processed using the material detection model, and the appearance information features are introduced to calculate the cosine distance corresponding to the target border information: ; Where, Indicates the The preset target data and The cosine distance between target bounding box information, Indicates the The appearance descriptor of the target bounding box information, Indicates the Preset target data successfully tracked The set of feature vectors of The default value is 100. Representing a collection Middle feature vectors, Represents cosine similarity.

[0031] Furthermore, the obtained Mahalanobis distance and cosine distance are weighted and combined, and the matching degree corresponding to the target feature is calculated using the weighted combination result: ; Where, Represents the weight coefficient.

[0032] S104, determining the target type of the target feature according to the matching degree, and acquiring corresponding detection conditions based on the target type; Furthermore, the step S104 specifically includes steps S1041 and S1042: S1041, calculating the matching degree using the Hungarian algorithm to construct a corresponding cost matrix, and constraining the cost matrix to obtain optimal allocation data corresponding to the matching degree; S1042: Adjust the cost matrix according to the optimal allocation data, and calculate the adjusted cost matrix to determine the target type of the target feature, and obtain corresponding detection conditions from a preset condition database according to the target type.

[0033] In the specific implementation, the Hungarian algorithm is used to calculate the matching degree, and the matching degree is converted into the corresponding cost matrix. Each row and each column in the cost matrix is ​​constrained. Specifically, the minimum value of each row is found, and the minimum value is subtracted from all elements of the row to obtain the row mark value. The minimum value of each column of the matrix after the row constraint is found, and the minimum value is subtracted from all elements of the column to obtain the column mark value, so as to complete the constraint of the cost matrix. Furthermore, all the marking values ​​in the cost matrix are covered with the least number of lines. Each row in the cost matrix is ​​traversed. If there is no independent row marking value in the row, the row is marked. Among all the marked rows, the column containing the column marking value is found. In the marked columns, the rows with independent row markings are found and marked. The cycle operation is repeated until no new rows or columns can be marked. If the number of covered values ​​is the same as the dimension of the cost matrix, several independent marking values ​​are found in the final matrix. The position of each marking value is the optimal allocation data. Specifically, the cost matrix is ​​adjusted according to the optimal allocation data, and the adjusted cost matrix is ​​calculated. The calculated result is compared with the preset type database to determine the target type of the target feature, and the corresponding detection condition is obtained in the preset condition database according to the target type.

[0034] S105: Determine whether the target feature meets the detection condition. If the target feature meets the detection condition, output the determination result of the target feature.

[0035] In specific implementation, it is determined whether the target feature meets the detection conditions obtained above. If the target feature meets the detection conditions, the target corresponding to the target feature is marked as meeting the requirements, and the corresponding judgment result is output.

[0036] In summary, the material detection method of the material receiving machine in the above embodiment of the present invention performs video capture on the preset area of ​​the material receiving machine, and processes the video data frame by frame to obtain the corresponding video frame sequence, and uses a frame-by-frame analysis method to improve the overall material detection accuracy; uses a deep learning model to perform sequence detection on the video frame sequence, and uses a material detection model to perform feature extraction on the target border information obtained by the sequence detection, and performs state prediction based on the target features to calculate the matching degree between the target features and the preset target data, and uses the matching degree method to realize rapid detection of materials, thereby improving the overall detection efficiency, and uses the matching degree to determine the corresponding detection conditions, and realizes efficient detection of materials through the detection conditions, thereby improving the accuracy of overall material detection. Example 2

[0037] Another aspect of the present invention also provides a material detection system for a material receiving machine, see Figure 2 , shown is a material detection system for a material receiving machine in a second embodiment of the present invention, comprising: The data processing module 11 is used to collect video data of a preset area on the material receiving machine in real time, and process the video data frame by frame to obtain a corresponding video frame sequence; Furthermore, the data processing module 11 is specifically configured to: Collecting video data of a preset area on the material receiving machine in real time, and processing the video data using a color conversion algorithm to obtain corresponding three-channel image data; The mask minimum circumscribed rectangle of the three-channel image data is calculated based on a preset mask algorithm, and frame alignment processing is performed using a temporal frame processing algorithm to obtain a corresponding video frame sequence.

[0038] A sequence detection module 12 is used to perform sequence detection on the video frame sequence using a preset deep learning model to obtain corresponding target frame information; Furthermore, the sequence detection module 12 is specifically configured to: Constructing a lightweight backbone network and replacing the backbone network of the YOLOv5 model with the lightweight backbone network to obtain an optimized YOLOv5 model; The optimized YOLOv5 model is input into a depthwise separable convolution algorithm to obtain a corresponding deep learning model, and the deep learning model is used to perform sequence detection on the video frame sequence to obtain corresponding target border information.

[0039] The state prediction module 13 is used to build a material detection model, extract the target features corresponding to the target frame information using the material detection model, and perform state prediction based on the target features to calculate the matching degree between the target features and the preset target data; Furthermore, the state prediction module 13 is specifically configured to: Processing the target border information using the material detection model to obtain a Mahalanobis distance and a cosine distance corresponding to the target border information; A weighted combination is performed according to the Mahalanobis distance and the cosine distance, and a matching degree corresponding to the target feature is calculated based on the weighted combination result.

[0040] a condition acquisition module 14, configured to determine a target type of the target feature according to the matching degree, and acquire corresponding detection conditions based on the target type; The condition judgment module 15 is used to judge whether the target feature meets the detection condition, and output the judgment result of the target feature if the target feature meets the detection condition.

[0041] Furthermore, the condition acquisition module 15 is specifically configured to: Calculating the matching degree using the Hungarian algorithm to construct a corresponding cost matrix, and constraining the cost matrix to obtain optimal allocation data corresponding to the matching degree; The cost matrix is ​​adjusted according to the optimal allocation data, and the adjusted cost matrix is ​​calculated to determine the target type of the target feature, and the corresponding detection condition is obtained from a preset condition database according to the target type.

[0042] The functions or operation steps implemented when the above modules are executed are substantially the same as those in the above method embodiments and will not be repeated here.

[0043] The material detection system for a material receiving machine provided in an embodiment of the present invention has the same implementation principle and technical effects as those in the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference can be made to the corresponding content in the aforementioned method embodiment. Example 3

[0044] The present invention also provides a computer, see Figure 3 , shown is a computer in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored in the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the material detection method of the material receiving machine mentioned above is implemented.

[0045] The memory 10 includes at least one type of storage medium, including flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 10 may be an internal storage unit of a computer, such as the computer's hard disk. In other embodiments, the memory 10 may also be an external storage device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the memory 10 may include both an internal storage unit of the computer and an external storage device. The memory 10 can be used not only to store application software installed in the computer and various types of data, but also to temporarily store data that has been output or is about to be output.

[0046] Among them, in some embodiments, the processor 20 can be an electronic control unit (Electronic Control Unit, abbreviated as ECU, also known as a vehicle computer), a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run the program code stored in the memory 10 or process data, such as executing access restriction programs.

[0047] It should be pointed out that Figure 3The structure shown does not constitute a limitation of the computer. In other embodiments, the computer may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0048] An embodiment of the present invention further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the material detection method of the material receiving machine as described above is implemented.

[0049] Those skilled in the art will appreciate that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device), or in conjunction with such instruction execution system, apparatus, or device. For purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by an instruction execution system, apparatus, or device, or in conjunction with such instruction execution system, apparatus, or device.

[0050] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting, or processing it in another suitable manner as necessary, and then storing it in a computer memory.

[0051] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the aforementioned embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following technologies known in the art may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0052] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0053] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A material detection method for a material receiving machine, characterized in that: The following steps are involved: Step 1: real-time acquisition of video data of a preset area on the material receiving machine, and frame-processing of the video data to obtain a corresponding video frame sequence; Step 2: Using a preset deep learning model to perform sequence detection on the video frame sequence to obtain corresponding target border information; Step 3: Construct a material detection model, and use the material detection model to extract the target features corresponding to the target frame information, and perform state prediction based on the target features to calculate the matching degree between the target features and the preset target data; Step 4: determining the target type of the target feature according to the matching degree, and obtaining corresponding detection conditions based on the target type; Step 5: Determine whether the target feature meets the detection condition. If the target feature meets the detection condition, output the determination result of the target feature.

2. The material detection method of the material receiving machine according to claim 1, characterized in that: The step one comprises: Collecting video data of a preset area on the material receiving machine in real time, and processing the video data using a color conversion algorithm to obtain corresponding three-channel image data; The mask minimum circumscribed rectangle of the three-channel image data is calculated based on a preset mask algorithm, and frame alignment processing is performed using a temporal frame processing algorithm to obtain a corresponding video frame sequence.

3. The material detection method of the material receiving machine according to claim 1, characterized in that: The second step includes: Constructing a lightweight backbone network and replacing the backbone network of the YOLOv5 model with the lightweight backbone network to obtain an optimized YOLOv5 model; The optimized YOLOv5 model is input into a depthwise separable convolution algorithm to obtain a corresponding deep learning model, and the deep learning model is used to perform sequence detection on the video frame sequence to obtain corresponding target border information.

4. The material detection method of the material receiving machine according to claim 1, characterized in that: The step three includes: Processing the target border information using the material detection model to obtain a Mahalanobis distance and a cosine distance corresponding to the target border information; A weighted combination is performed according to the Mahalanobis distance and the cosine distance, and a matching degree corresponding to the target feature is calculated based on the weighted combination result.

5. The material detection method of the material receiving machine according to claim 1, characterized in that: The fourth step includes: Calculating the matching degree using the Hungarian algorithm to construct a corresponding cost matrix, and constraining the cost matrix to obtain optimal allocation data corresponding to the matching degree; The cost matrix is ​​adjusted according to the optimal allocation data, and the adjusted cost matrix is ​​calculated to determine the target type of the target feature, and the corresponding detection condition is obtained from a preset condition database according to the target type.

6. A material detection system for a material receiving machine, characterized in that: include: A data processing module is used to collect video data of a preset area on the material receiving machine in real time, and process the video data frame by frame to obtain a corresponding video frame sequence; A sequence detection module is used to perform sequence detection on the video frame sequence using a preset deep learning model to obtain corresponding target frame information; A state prediction module is used to build a material detection model, and use the material detection model to extract the target features corresponding to the target frame information, and perform state prediction based on the target features to calculate the matching degree between the target features and the preset target data; a condition acquisition module, configured to determine a target type of the target feature according to the matching degree, and acquire corresponding detection conditions based on the target type; The condition judgment module is used to judge whether the target feature meets the detection condition, and if the target feature meets the detection condition, output the judgment result of the target feature.

7. The material detection system of the material receiving machine according to claim 6, characterized in that: The data processing module is specifically used for: Collecting video data of a preset area on the material receiving machine in real time, and processing the video data using a color conversion algorithm to obtain corresponding three-channel image data; The mask minimum circumscribed rectangle of the three-channel image data is calculated based on a preset mask algorithm, and frame alignment processing is performed using a temporal frame processing algorithm to obtain a corresponding video frame sequence.

8. The material detection system of the material receiving machine according to claim 6, characterized in that: The sequence detection module is specifically used for: Constructing a lightweight backbone network and replacing the backbone network of the YOLOv5 model with the lightweight backbone network to obtain an optimized YOLOv5 model; The optimized YOLOv5 model is input into a depthwise separable convolution algorithm to obtain a corresponding deep learning model, and the deep learning model is used to perform sequence detection on the video frame sequence to obtain corresponding target border information.

9. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the material detection method of the material receiving machine as described in any one of claims 1 to 5 is implemented.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the material detection method of the material receiving machine as described in any one of claims 1 to 5 is implemented.

Images