Material detection method and system for material stacking area, medium and product

Abstract

The invention provides a material detection method and system for a material stacking area, a medium and a product, and relates to the technical field of image equipment. Firstly, a monitoring image of a test paper storage area is collected and differentiated definition processing is carried out, so that a test paper bag area can be highlighted in the image because the definition is higher than that of a non-test paper bag area, different parts in a test paper bag are differentiated in definition, and the contour of the test paper bag is the highest; since the distinction degree of test paper bags and the distinction degree of boundaries need to be increased and different resolution sampling is adopted for the contour of the test paper bags according to the structural complexity, the resolution of a simple-structure area is improved and the contour can be caught more finely, and the resolution is reduced and resource consumption and interference are reduced because the distinction of a complex area is obvious. The matching of the two optimizes the contour information acquisition, and reduces the risks of missing detection and false detection.

Description

Technical Field

[0001] The present application relates to the technical field of image devices, and particularly to a method, a system, a medium, and a product for detecting materials in a material stacking area. Background Art

[0002] In the field of education, products printed in batches such as test papers and exercise books have clear distribution requirements. They need to be packaged separately according to various factors such as different regions, schools, and specific classes to ensure accurate supply to corresponding usage scenarios. For example, primary schools of different scales have different numbers of classes, and the number of students in each class is also different. For example, Primary School A has a total of 4 classes, with 28 students in Class A1, 32 students in Class A2, 31 students in Class A3, and 25 students in Class A4. Primary School B has a total of 3 classes, with 22 students in Class B1, 32 students in Class B2, and 28 students in Class B3. Moreover, in many cases, these test papers also need to be sealed and transported to designated places such as examination rooms. Therefore, after printing, they need to be strictly packaged, sealed, and transported according to the class situation. The sealed and rechecked test paper bags and test papers are stored in a designated area, and a detection system monitors the materials in the designated area in real time to prevent the loss or omission of the rechecked test paper bags.

[0003] However, in the related art, during the process of detecting these test paper bags, since all test papers are packaged using kraft paper of the same specification, the following technical difficulties are caused: 1) There are no significant visual feature differences between the test paper bags. Since the same kraft paper makes all test paper bags show a uniform off-white color, and the same packaging specification results in the test paper bags having nearly identical volumes and shapes, it is difficult for an image recognition system to extract reliable distinguishing information from the appearance features; 2) When multiple batches of test papers are stacked, the boundaries between the test paper bags are blurred, and coupled with the interference of light and shadows, the image segmentation algorithm based on contour detection has poor effects and cannot accurately define the spatial range of each test paper bag; these problems lead to the inability of the related image recognition technology to accurately identify the test paper batches, and there are risks of missed detection and false detection. In scenarios such as educational examinations where the tolerance for missed detection and false detection is relatively low, the impact is more obvious. Summary of the Invention

[0004] The present application provides a method, a system, a medium, and a product for detecting materials in a material stacking area, which are used to reduce the risks of missed detection and false detection.

[0005] In a first aspect, the present application provides a method for detecting materials in a material stacking area, including: collecting monitoring images of the test paper storage area, and performing differential clarity processing on the monitoring images to obtain a sequence of feature maps, where the clarity of the image in the test paper bag area is higher than that in the non-test paper bag area. Within the test paper bag area, the clarity of the test paper bag contour, other areas, and the test paper bag body area decreases in sequence; increasing the resolution sampling for the simple structure area of the test paper bag contour and decreasing the resolution sampling for the complex structure area of the test paper bag contour; in the case where it is detected that the position of the test paper bag has changed, performing spatial registration calculation on the sequence of feature maps from before the previous position change to the current moment with a dynamic reference model to obtain the displacement amounts of different test paper bags; the dynamic reference model is a three-dimensional space mapping established for the initial stacking state when the test papers are put into storage; determining the target test paper bags to be reduced or increased according to the displacement amounts of the test paper bags; collecting the interaction data between the personnel activity trajectory and the test paper bag area, and performing matching analysis on the interaction data with a preset operation procedure; when the interaction data conforms to the preset operation procedure, determining the corresponding behavior as a normal event; when the interaction data does not conform to the preset work procedure, determining the corresponding behavior as an abnormal event; judging the target test paper bags corresponding to the normal events with the preset operation procedure; when the test paper bags to be stored and retrieved specified by the preset operation procedure do not correspond to the target test paper bags, changing the corresponding normal event to an abnormal event; marking the abnormal events.

[0006] By adopting the above technical solution, first, the monitoring images of the test paper storage area are collected and differential clarity processing is performed. This operation enables the test paper bag area to stand out in the image because its clarity is higher than that in the non-test paper bag area, and there are also distinctions in the clarity of different parts inside the test paper bag. The clarity of the test paper bag contour is the highest because it is necessary to increase the distinctiveness of the test paper bag and the distinctiveness of the boundary. At the same time, then different resolution samplings are taken for the test paper bag contour according to the complexity of the structure. Increasing the resolution for the simple structure area can capture the contour more precisely, and for the complex area, since the distinction is already obvious, reducing the resolution can reduce resource consumption and interference. The two cooperate with each other to optimize the acquisition of contour information. Then, the dynamic reference model is used to perform spatial registration calculation of the displacement amount, and the current position change of the test paper bag is accurately compared based on the three-dimensional space mapping of the initial state when the test papers are put into storage. Then, through a series of rigorous judgment steps, the target test paper bags are determined, overcoming problems such as similar appearance and blurred boundaries of the test paper bags, and finally effectively improving the ability to accurately identify the test paper batches and reducing the risks of missed inspection and misinspection.

[0007] In some embodiments in combination with some embodiments of the first aspect, the step of determining the target test paper bags to be reduced or increased according to the displacement amount of the test paper bags specifically includes: determining the relative positional relationship between the test paper bag areas; establishing the association strength between different test paper bag stacks, where the association strength decays as the distance between the test paper bag stacks increases; the test paper bag stacks are all the test paper bags within the test paper bag areas; determining the change amount of the distance between the test paper bags to be reduced or increased in the test paper bag stack and the test paper bag stack to which they belong; determining the group behavior consistency index according to all the change amounts; if the change amount of the distance between the test paper bags to be reduced or increased in the test paper bag stack and the test paper bag stack to which they belong is greater than the preset threshold and is outside the corresponding group behavior consistency index; determining them as the target test paper bags.

[0008] By adopting the above technical solution, the relative positional relationship between the test paper bag areas is determined, which can clarify the layout of each test paper bag stack in space. Then, the association strength between different test paper bag stacks is established, and the association strength decays as the distance increases, so that the closeness of the association between each test paper bag stack can be measured according to the distance. Subsequently, the change amount of the distance between the test paper bags to be reduced or increased in the test paper bag stack and the test paper bag stack to which they belong is analyzed, and the group behavior consistency index is combined for comprehensive judgment. When there are addition or subtraction changes in the test paper bag stack and meet the corresponding conditions, they are determined as the target test paper bags, deeply considering the overall relevance and changes of the test paper bag stack, so as to distinguish whether it is a reasonable change during the normal handling and stacking process or a real batch confusion caused by improper handling, improving the accuracy of judging whether the test paper batch is abnormal.

[0009] In some embodiments in combination with some embodiments of the first aspect, after the step of collecting the interaction data between the personnel activity trajectory and the test paper bag area and performing matching analysis on the interaction data with the preset operating procedures, the method further includes: obtaining the type identifier of the test paper bag, where the type identifier is used to group the test paper bags together, and the type identifier includes one or more of school, class, and printed product subject information; grouping the collected personnel activity trajectories by the type identifier; randomly selecting two collected personnel activity trajectories in the same group of collected personnel activity trajectories and judging the similarity; if the similarity is lower than the similarity threshold, then extracting the avoidance behavior characteristics from the collected personnel activity trajectory with a later time sequence; identifying the obstacle test paper bags corresponding to the avoidance behavior characteristics; obtaining the first test paper bag corresponding to the collected personnel activity trajectory with an earlier time sequence and the second test paper bag corresponding to the collected personnel activity trajectory with a later time sequence; judging whether the obstacle test paper bags meet the requirements according to the preset operating procedures for being carried after the first test paper bag and before the second test paper bag; if not all meet, then determining the corresponding behaviors of the obstacle test paper bags, the first test paper bag, and the second test paper bag as abnormal events.

[0010] By adopting the above technical solution, the type identifier of the test paper bag is first obtained, and the activity trajectories of the collectors are grouped based on this, so that the operations of the personnel related to the test papers of the same type can be grouped together, which is convenient for targeted analysis. Among the activity trajectories of the collectors in the same group, two trajectories are randomly selected to judge the similarity. If it is lower than the threshold, the avoidance behavior characteristics are extracted, and the obstacle test paper bag is identified through this characteristic. Then, combined with the test paper bags corresponding to different time sequences before and after, it is judged whether the handling order complies with the preset operation procedures according to the preset operation procedures. Starting from the perspective of the activity trajectories of the personnel, the possible errors in the test paper distribution and other links due to incorrect order are deeply explored, and the incorrect behaviors with non-standard time sequences are found.

[0011] Combined with some embodiments of the first aspect, in some embodiments, the step of extracting the avoidance behavior characteristics from the activity trajectories of the collectors with later time sequences specifically includes: respectively obtaining the position sequences and speed sequences of the two activity trajectories of the collectors; calculating the curvature values corresponding to the position sequences of the two activity trajectories of the collectors; in the activity trajectories of the collectors with later time sequences, when the curvature value is greater than the first preset threshold and the corresponding speed value is less than the second preset threshold, determining the candidate avoidance positions; judging whether the activity trajectories of the collectors with earlier time sequences satisfy that the curvature value is greater than the first preset threshold and the speed value is less than the second preset threshold at the corresponding spatial positions; in the case of satisfaction, determining the candidate avoidance positions as the avoidance behavior characteristics.

[0012] By adopting the above technical solution, the position and speed sequences of the two activity trajectories of the collectors are obtained. The curvature value corresponding to the position sequence can reflect the degree of bending of the trajectory, and the speed value reflects the speed of movement. The candidate avoidance positions with large curvature and small speed in the trajectory with later time sequence are screened out by setting thresholds, and then the corresponding spatial position conditions of the trajectory with earlier time sequence are compared, and the two confirm each other to determine the avoidance behavior characteristics.

[0013] Combined with some embodiments of the first aspect, in some embodiments, the step of identifying the obstacle test paper bag corresponding to the avoidance behavior characteristics specifically includes: constructing a search area for the avoidance behavior characteristics; determining the target area closest to the search area in the preset operation procedures; determining the test paper bag corresponding to the target area as the obstacle test paper bag.

[0014] By adopting the above technical solution, a search area is constructed based on the determined avoidance behavior characteristics, and then the target area closest to it is determined according to the preset operation procedures, and then the corresponding test paper bag is locked as the obstacle test paper bag. Starting from the personnel avoidance behavior, the obstacle test paper bag that may affect the operation process is accurately associated.

[0015] In some embodiments in combination with some embodiments of the first aspect, the step of performing differential clarity processing on the monitoring image to obtain a sequence of feature maps specifically includes: extracting the boundary of the test paper bag in the monitoring image to obtain a preliminary contour of the test paper bag; determining whether the preliminary contour of the test paper bag meets the requirements of a simple region; if it meets the requirements, constructing a multi-scale feature pyramid, performing feature extraction on the contour of the test paper bag at different scales to obtain local fine features; and fusing the local fine features with the preliminary contour of the test paper bag to obtain the contour structure of the test paper bag.

[0016] By adopting the above technical solution, first extract the boundary of the test paper bag in the monitoring image to obtain a preliminary contour, and after determining whether it meets the requirements of a simple region, for those that meet the requirements, construct a multi-scale feature pyramid to extract features at different scales, and then fuse them to obtain a more perfect contour structure of the test paper bag. Through such multi-step feature extraction and fusion, the presentation effect of the contour of the test paper bag can be strengthened, key features can be highlighted, the recognition problem caused by the similar appearance of test paper bags in the image can be overcome, and it can better assist in the subsequent accurate analysis and judgment of the test paper bag.

[0017] In some embodiments in combination with some embodiments of the first aspect, after the step of marking an abnormal event, the method further includes: associating the corresponding batch identification information, and recording the operator's identity information, operation timestamp, and spatial position coordinates.

[0018] By adopting the above technical solution, after marking the abnormal event, associate the corresponding batch identification information, and record the operator's identity, operation timestamp, and spatial position coordinates. In this way, the detailed background information of the abnormal event can be retained in all aspects, which is convenient for tracing the root cause of the abnormality and accurately reviewing the event process subsequently.

[0019] In a second aspect, the present application provides a material detection system for a material stacking area. The material detection system for the material stacking area includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the material detection system for the material stacking area to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0020] In a third aspect, the present application provides a computer program product containing instructions. When the computer program product runs on the material detection system for the material stacking area, it enables the material detection system for the material stacking area to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] Fourthly, the present application provides a computer-readable storage medium, including instructions, which, when running on a material detection system in a material stacking area, cause the material detection system in the material stacking area to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. First, monitor images of the test paper storage area are collected and processed for differential clarity. This operation makes the test paper bag area stand out in the image because its clarity is higher than that of non-test paper bag areas, and there are also distinctions in the clarity of different parts inside the test paper bag. The outline of the test paper bag is the highest because it is necessary to increase the distinctiveness of the test paper bag and the distinctiveness of the boundary. At the same time, different resolution samplings are taken for the outline of the test paper bag according to the complexity of the structure. Increasing the resolution in simple-structured areas can capture the outline more precisely, and reducing the resolution in complex areas can reduce resource consumption and interference because the distinctions are already obvious. The two cooperate with each other to optimize the acquisition of outline information. Then, a dynamic reference model is used to perform spatial registration to calculate the displacement amount, and the current position change of the test paper bag is accurately compared based on the three-dimensional space mapping of the initial state of test paper warehousing. Then, through a series of rigorous judgment steps, the target test paper bag is determined, overcoming problems such as similar appearances and blurred boundaries of test paper bags, and finally effectively improving the ability to accurately identify test paper batches and reducing the risks of missed inspection and misjudgment.

[0023] 2. Determine the relative position relationship between test paper bag areas, which can clarify the layout of each test paper bag stack in space. Then, establish the association strength between different test paper bag stacks, and the association strength decays as the distance increases. In this way, the closeness of the association between each test paper bag stack can be measured according to the distance. Subsequently, analyze the change amount of the distance between the test paper bags reduced or increased in the test paper bag stack and the belonging test paper bag stack, and comprehensively judge in combination with the group behavior consistency index. When there are changes in the test paper bag stack and meet the corresponding conditions, it is determined as the target test paper bag, deeply considering the overall relevance and changes of the test paper bag stack, so as to distinguish whether it is a reasonable change during normal handling and stacking processes or a real batch chaos caused by improper handling, improving the accuracy of judging whether the test paper batch is abnormal.

[0024] 3. First, obtain the type identifier of the test paper bag, and group the collected personnel activity trajectories based on this, so that the personnel operations related to the same type of test paper can be grouped together, which is convenient for targeted analysis. In the collected personnel activity trajectories of the same group, randomly select two trajectories to judge the similarity. If it is lower than the threshold, extract the avoidance behavior characteristics, and identify the obstacle test paper bag through this characteristic. Then, combine the test paper bags corresponding to different time sequences before and after, and judge whether the handling order complies with the preset operation procedures. Starting from the perspective of personnel activity trajectories, deeply explore possible errors in the test paper distribution and other links due to incorrect sequences, and find out incorrect behaviors with non-standard time sequences. Brief Description of the Drawings

[0025] Figure 1 is a schematic flowchart of a method for detecting materials in a material stacking area in an embodiment of the present application; Figure 2 is another schematic flowchart of a method for detecting materials in a material stacking area in an embodiment of the present application; Figure 3 is another schematic flowchart of a method for detecting materials in a material stacking area in an embodiment of the present application; Figure 4 is a schematic diagram of an exemplary hardware structure of a material detection system in a material stacking area in an embodiment of the present application. Detailed Description of the Embodiments

[0026] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "an", "the", "above-mentioned", "the foregoing", "this" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more of the listed items.

[0027] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as implying or indicating relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0028] Please refer to Figure 1 , Figure 1 is a schematic flowchart of a method for detecting materials in a material stacking area in an embodiment of the present application; S101. Collect monitoring images of the test paper storage area, and perform differential clarity processing on the monitoring images to obtain a sequence of feature maps, where the image clarity of the test paper bag area is higher than that of the non-test paper bag area. Within the test paper bag area, the clarity of the test paper bag contour, other areas, and the test paper bag body area decreases in turn; increase the resolution sampling for the simple structure area of the test paper bag contour and decrease the resolution sampling for the complex structure area of the test paper bag contour; Among them, the test paper bag area refers to the part of the image area where the bag storing the test papers is located, and it is the object of key attention and analysis. The non-test paper bag area refers to other areas except the test paper bag area. The test paper bag contour is the outer edge shape line part presented by the test paper bag in the image, and the processing of its clarity is crucial for subsequent recognition. The other area refers to the surrounding relevant image area except the test paper bag and its internal parts. The test paper bag body area is the image area corresponding to the internal main part of the test paper bag itself except the contour.

[0029] Specifically, first, a camera is used to collect the monitoring image of the test paper storage area, and then an image algorithm is used to perform differential clarity processing on the collected monitoring image.

[0030] During the processing, the clarity of the image in the test paper bag area is made higher than that in the non-test paper bag area, and within the test paper bag area, the clarity is further refined so that the clarity of the test paper bag contour is the highest, because its clear contour helps to accurately distinguish different test paper bags. Then comes the other area, because the other area is used to judge whether there is an interaction behavior, and finally the test paper bag body area. After that, for the test paper bag contour, judge whether its structure is simple or complex. For the simple structure area, increase the resolution sampling, so that the contour details can be captured more precisely; while for the complex structure area, since the contour features are relatively obvious itself, reduce the resolution sampling, which can not only reduce resource consumption but also avoid excessive interference, and finally obtain a sequence of feature maps, providing a valuable image data basis for the subsequent steps.

[0031] In some specific embodiments, an image recognition algorithm is used to analyze the collected monitoring image to locate the position of the test paper bag in the image. Subsequently, according to a preset specific interval range, the identified test paper bag is completely framed, and in this way, this framed part is determined as the test paper bag area, and the remaining part of the monitoring image except the test paper bag area naturally belongs to the non-test paper bag area. Then, for the determined test paper bag area, the image recognition algorithm is used again to further identify the test paper bag contour therein. The part covered by the test paper bag contour is clearly defined as the test paper bag body area. Correspondingly, the other parts in the test paper bag area except the test paper bag body area are identified as the other area. After the accurate division of the above-mentioned areas is completed, according to the preset requirements, that is, the clarity of the image in the test paper bag area is higher than that in the non-test paper bag area, and within the test paper bag area, the clarity of the test paper bag contour is higher than that of the test paper bag body area, and the clarity of the test paper bag body area is higher than that of the other area, corresponding image processing technologies are used to perform differential clarity processing on different areas. No limitation is made here.

[0032] In some specific embodiments, the curvature change of the contour curve is calculated, and the number of points with frequent curvature changes (i.e., the curvature value fluctuates greatly multiple times within a short distance) and the number of curvature mutations (such as suddenly changing from gentle to sharply curved) are counted. If these statistical values are lower than a set threshold (which can be determined by testing and analyzing a large number of samples), the shape is considered regular and the structure is determined to be simple; if it exceeds the threshold, it is determined to be a complex structure. In some other embodiments, parameters are taken according to the curvature change situation, and the curvature change situation is inversely proportional to the parameter, and the parameter is used to limit the clarity and is directly proportional to the clarity.

[0033] In some specific embodiments, since the test paper bags are stacked together, a segmentation operation is performed on the contour of the test paper bag, aiming to reasonably divide the contour of the test paper bag in this way. Viewing the contour of the test paper bag from a horizontal perspective, it presents several relatively uniform shapes, mostly rectangular. When trying to approximately represent the contour of the test paper bag with basic geometric figures, if only a few rectangles are needed to relatively accurately fit it, it means that the structure of the contour of the test paper bag is relatively simple. However, if a large number of rectangles are needed, and even other various geometric figures need to be combined to approximately represent the contour of the test paper bag, it indicates that the structure of the contour of the test paper bag is relatively complex. This shows that there are many concavities, convexities, bends and irregular changes in its contour, and the stacking is uneven. At the same time, corresponding parameters are obtained according to the number of rectangles or other geometric figures used, and there is an inverse relationship between the curvature change situation of the contour and the parameter, and the parameter is used to limit the clarity and is directly proportional to the clarity.

[0034] In some specific embodiments, for the simple region of the test paper bag contour, it specifically includes: S1011. Extract the boundary of the test paper bag in the monitoring image to obtain a preliminary test paper bag contour; S1012. Determine whether the preliminary test paper bag contour meets the requirements of the simple region; S1013. If it meets the requirements, construct a multi-scale feature pyramid, extract features of different scales from the test paper bag contour to obtain local fine features; S1014. Fuse the local fine features with the preliminary test paper bag contour to obtain the test paper bag contour structure.

[0035] It can be seen that, first, the boundary of the test paper bag in the monitored image is extracted to obtain a preliminary contour. After determining whether it meets the requirements of a simple region, for those that meet the requirements, a multi-scale feature pyramid is constructed to extract features at different scales, and then fused to obtain a more perfect contour structure of the test paper bag. Through such multi-step feature extraction and fusion, the presentation effect of the test paper bag contour can be strengthened, key features can be highlighted, the recognition problem caused by the similar appearance of test paper bags in the image can be overcome, and it can better assist in the subsequent accurate analysis and judgment of the test paper bag.

[0036] S102. In the case where the position of the test paper bag is detected to have changed, perform spatial registration calculation on the feature map sequence from before the previous position change to the current moment with the dynamic reference model to obtain the displacement amounts of different test paper bags; the dynamic reference model is a three-dimensional space mapping established for the initial stacking state when the test papers are stored in the warehouse. Specifically, in the scenario where the test paper storage area is continuously monitored and the position of the test paper bag is detected to have changed through image analysis, the feature map sequence from before the previous position change to the current moment is used as the data to be analyzed, and then spatial registration calculation is performed with the pre-constructed dynamic reference model. This dynamic reference model is based on the initial stacking state when the test papers are stored in the warehouse, and details such as the initial spatial positions of each test paper bag and the layout relationship between them are recorded through three-dimensional modeling and other technologies. When performing spatial registration calculation, using the image space transformation algorithm, the images of each test paper bag in the feature map sequence are precisely matched and aligned with the corresponding initial states in the dynamic reference model. After a series of complex operations such as coordinate transformation and similarity calculation, the displacement amounts of different test paper bags relative to the initial state in terms of spatial position are finally obtained, providing a basis for subsequent judgment of the increase or decrease of test paper bags, etc.

[0037] In some embodiments, first, key feature points of the images of each test paper bag in the feature map sequence are extracted, such as contour corners, specific identification positions, etc., and at the same time, the corresponding initial feature points are also marked in the dynamic reference model; second, a feature point matching algorithm, such as the SIFT (Scale-Invariant Feature Transform) algorithm, is used to find the corresponding feature point pairs between the feature map sequence and the dynamic reference model; third, according to the successfully matched feature point pairs, calculation methods such as the least squares method are used to determine the spatial coordinate changes of each test paper bag image relative to the initial state, and then the displacement amounts are obtained, which are not limited herein.

[0038] S103. Determine the target test paper bags to be reduced or increased according to the displacement amounts of the test paper bags. In some embodiments, different displacement amount thresholds are set, and the displacement amount of each test paper bag is compared with the threshold, and the test papers with displacement amounts exceeding the threshold are screened out and determined as the target test paper bags.

[0039] S104. Collect the interaction data between the personnel's activity trajectories and the test paper bag area, and perform matching analysis on the interaction data with the preset operating procedures; Among them, the personnel's activity trajectories refer to the action route records left by relevant staff during various operations in the test paper storage area, which can usually be obtained through positioning devices, surveillance video analysis, etc., and reflect the activities of personnel. The interaction data of the test paper bag area refers to various data information generated during operations related to the test paper bag area such as approaching, contacting, and handling, such as contact time, operation sequence, etc., which is an important basis for analyzing whether the operations are compliant. The preset operating procedures include which personnel have the corresponding permissions and can have interaction behaviors such as handling, viewing, storing, and retrieving with which specific test paper bags, while other irrelevant personnel are prohibited from having interactions with them, so as to ensure the standardization, safety, and confidentiality of operations related to test paper bags.

[0040] In some embodiments, use the video surveillance system to analyze the video footage through image recognition algorithms to extract interaction data such as the action behaviors and staying time of personnel in the test paper bag area; organize the preset operating procedures into a detailed table form to clarify the requirements of each operation step and the sequence; manually or with the help of data analysis software, compare the extracted interaction data with the requirements in the table one by one to determine whether they match, and then complete the analysis.

[0041] S105. When the interaction data conforms to the preset operating procedures, determine the corresponding behavior as a normal event; when the interaction data does not conform to the preset operating procedures, determine the corresponding behavior as an abnormal event; Specifically, after comparison, it is found that the personnel operation behaviors reflected in the interaction data, such as the operation sequence, action specifications, and test paper bags involved, all correspond one by one to the content stipulated in the preset operating procedures without any violation of the rules, then the corresponding behavior is determined as a normal event. On the contrary, if it is found during the comparison process that the operations do not conform to the preset operating procedures, then the corresponding behavior is determined as an abnormal event, so as to take targeted measures subsequently, such as further investigating and correcting the abnormal event.

[0042] S106. Judge the target test paper bag corresponding to the normal event against the preset operating procedures; Among them, the preset operating procedures include the target test paper bags, quantities, target areas, etc. that the corresponding personnel should store in a certain batch, which are used as an important reference basis for measuring whether the actual operations are completely compliant.

[0043] In some specific embodiments, key elements regarding the access of test paper bags in the preset working procedures are sorted out, such as storage location requirements, target test paper bags, quantity, operators, etc., to form a detailed comparison list; for each target test paper bag corresponding to a normal event, the relevant information thereof is extracted and sorted into a data list in the same format; the two lists are compared item by item to check whether they are completely consistent. If they are all consistent, it is determined to meet the requirements. If there are any inconsistencies, it is determined that there may be problems, which is not limited herein.

[0044] S107. When the test paper bags accessed as specified in the preset working procedures do not correspond to the target test paper bags, change the corresponding normal event to an abnormal event. Specifically, after determining the target test paper bags corresponding to the normal events and comparing them with the preset working procedures, if it is found that the test paper bags accessed as specified in the preset working procedures do not correspond to the actual target test paper bags. For example, if the procedure requires accessing test paper bags of batch A, but the actual displacement and other situations involve test paper bags of batch B, it means that there may actually be problems with the previously determined normal events. Although the operation seemingly complies with some procedures on the surface, there are deviations in the key test paper bag objects. At this time, the corresponding normal event needs to be changed to an abnormal event so as to conduct in-depth investigations and corrections on such operations that seemingly comply with the regulations but actually have potential hazards in the follow-up, and to avoid causing adverse effects on the test paper management work.

[0045] S108. Mark the abnormal events.

[0046] In some embodiments, it further includes: S109. Associate the corresponding batch identification information, and record the operator's identity information, operation timestamp, and spatial position coordinates It can be seen that first, the monitoring images of the test paper storage area are collected and processed for differential clarity. This operation makes the test paper bag area stand out in the image because its clarity is higher than that of the non-test paper bag area, and there are also distinctions in the clarity of different parts inside the test paper bag. The outline of the test paper bag is the highest because it is necessary to increase the distinguishability of the test paper bag and the distinguishability of the boundary. At the same time, different resolutions are sampled for the test paper bag outline according to the structural complexity. Improving the resolution in the simple-structured area can capture the outline more precisely, and reducing the resolution in the complex area can reduce resource consumption and interference because the differences are already obvious. The two cooperate with each other to optimize the acquisition of outline information. Then, a dynamic reference model is used to perform spatial registration calculations for the displacement amount, and the current position change of the test paper bag is accurately compared based on the three-dimensional space mapping of the initial state of the test paper warehousing. Through a series of rigorous judgment steps, the target test paper bag is determined, overcoming problems such as similar appearances and blurred boundaries of the test paper bags, and finally effectively improving the ability to accurately identify the test paper batches and reducing the risks of missed inspections and misinspections.

[0047] The above embodiments solve the thorny problem of the risk of missed inspection and misjudgment due to the extremely small distinguishability and extremely close contours between the test paper bags. However, in the actual use process, test paper bags of different batches may be similar in appearance, specifications, etc. Relying solely on detecting the batch of the target test paper bag for judgment may lead to misjudgment. Maybe it's just because the two batches of test paper bags look very similar, and visual confusion occurs during normal handling and stacking, but there is actually no real batch confusion, or maybe there is batch confusion. For example, test paper bags printed by the same examination institution for two different schools are extremely similar in color, size, and basic markings. When it is found during monitoring that a certain test paper bag appears in the "other batch", and the "other batch" just happens to have fewer test paper bags, it may just be that this batch of test papers has just been moved out while new test papers have been brought in, rather than improper handling causing it to deviate from its original batch. For example, it may also be due to handling errors that cause it to deviate from its original batch. To sum up, based on two similar test paper stacks, a "reasonable change" is made (one test paper stack increases and one test paper stack decreases). How to distinguish whether an error has occurred or it is a normal handling behavior?

[0048] Please refer to Figure 2 , Figure 2 which is another process schematic diagram of the material detection method for the material stacking area in the embodiments of the present application; In some specific embodiments, step S103 specifically includes: S201. Determine the relative positional relationship between the test paper bag areas; The relative positional relationship refers to the mutual association between different test paper bag areas in terms of spatial orientation, distance, etc. It is a reference and comparison information used to describe the positions of each test paper bag area. For example, the test paper bag area A is 2 meters away from area B, and this is a relative positional relationship.

[0049] In some embodiments, just input the coordinate information corresponding to all the test paper bag areas into the system one by one, and then, based on these coordinate data, through the corresponding spatial position analysis algorithm for calculation and processing, the specific relative positional relationship between each test paper bag area can be obtained.

[0050] S202. Establish the association strength between different test paper bag stacks, where the association strength decays as the distance between the test paper bag stacks increases; the test paper bag stack is all the test paper bags within the test paper bag area; Specifically, for every two stacks of test paper bags, the association strength is established according to the distance between them. For stacks of test paper bags that are closer, since the possibility of influencing each other during operations such as handling and stacking is greater, the association strength is set relatively high; while as the distance between them increases, the probability of mutual influence decreases, and the association strength also decays accordingly. This provides a basis for subsequent judgment on whether the change of test paper bags is reasonable.

[0051] S203. Determine the change amount of the distance between the test paper bags reduced or increased in the stack of test paper bags and the stack of test paper bags to which they belong; Specifically, first accurately identify which stack of test paper bags has test paper bags reduced or increased, track the current spatial positions of these test paper bags with quantity changes, and then combine the initial position information of these test paper bags in the stack of test paper bags to which they belong, and use the spatial distance calculation algorithm to calculate the change amount of the distance of these test paper bags relative to the stack of test paper bags to which they belong. In this way, the specific position change situation of the test paper bags is obtained, providing detailed data support for subsequent comprehensive judgment.

[0052] S204. Determine the group behavior consistency index according to all the change amounts; Among them, the group behavior consistency index is a quantitative index that comprehensively reflects whether the overall behavior of multiple test paper bags conforms to normal rules and is consistent during the change process. It comprehensively considers various factors such as the position changes of different test paper bags and the association situation of the stacks of test paper bags to which they belong, and is used to measure whether the change of the entire group of test paper bags is in a reasonable and normal state. For example, under normal handling operations, the changes of test paper bags in the same batch are often more synchronous and regular, and the group behavior consistency index will be higher, otherwise it will be lower.

[0053] In some embodiments, for the test paper bags in each stack of test paper bags, calculate the similarity of the change amounts of the distances between every two of them. Methods such as Euclidean distance and cosine similarity can be used to measure. Integrate the association strength and similarity calculation index. According to the previously established association strength between the stacks of test paper bags, assign corresponding weights to the similarity calculation results of each stack of test paper bags. The higher the association strength, the greater the weight. Then, summarize the weighted similarity results of all stacks of test paper bags to obtain the group behavior consistency index.

[0054] S205. If the change amount of the distance between the test paper bags reduced or increased in the stack of test paper bags and the stack of test paper bags to which they belong is greater than the preset threshold and is outside the corresponding group behavior consistency index; determine them as target test paper bags.

[0055] Specifically, after determining the group behavior consistency index based on all the transformation amounts, it is necessary to judge which changes in the test paper bags are truly abnormal according to the set criteria, and then this step is executed. Compare the transformation amount of the distance between the test paper bags reduced or increased in the test paper bag stack and the corresponding test paper bag stack with the preset threshold, and at the same time check whether it deviates from the requirements of the corresponding group behavior consistency index. If the transformation amount of the distance of a certain test paper bag is greater than the preset threshold, it proves that it is not a temporary placement, and its change does not conform to the normal change law reflected by the group behavior consistency index, which means that the change of this test paper bag is likely not caused by normal operations such as handling and stacking, but there are abnormal situations such as handling errors and misplacement. At this time, this test paper bag is determined as the target test paper bag.

[0056] It can be seen that determining the relative position relationship between the test paper bag areas can clarify the layout of each test paper bag stack in space. Then establish the association strength between different test paper bag stacks, and the association strength decays as the distance increases, so that the association tightness between each test paper bag stack can be measured according to the distance. Subsequently, analyze the transformation amount of the distance between the test paper bags reduced or increased in the test paper bag stack, and make a comprehensive judgment in combination with the group behavior consistency index. When there are increase or decrease changes in the test paper bag stack and meet the corresponding conditions, it is determined as the target test paper bag, and deeply consider the overall relevance and change situation of the test paper bag stack, so as to distinguish whether it is a reasonable change in the normal handling and stacking process or a real batch confusion caused by improper handling, improving the accuracy of judging whether the test paper batch is abnormal.

[0057] It should be noted that the above embodiments are applicable to the scenario where a test paper stack is increased or decreased multiple times. For example, in a large-scale educational examination test paper storage library, there is a test paper stack dedicated to storing the college entrance examination test papers of a certain subject. Due to the differences in the examination arrangements in different examination areas for this subject, it is necessary to frequently transfer test papers from this test paper stack to some examination areas, and at the same time, newly printed spare test papers will be supplemented, so that the number of test paper bags in this test paper stack will decrease or increase multiple times.

[0058] The above embodiments solve the thorny problem of the risk of missed detection and false detection caused by the extremely small distinguishability and extremely close contours between test paper bags. However, in the actual use process, even if the personnel collecting the relevant information of the test paper batch are correct and the set test paper taking is without deviation, there are still deficiencies in the specific scenario of educational examination test paper batch monitoring. The reason is that the subsequent test paper distribution link is operated in a certain order. If everything else except the distribution order is accurate, but only the order is wrong, then there will still be corresponding risks for the back end.

[0059] Please refer to Figure 3 , Figure 3 which is another process schematic diagram of the material detection method in the material stacking area in the embodiment of the present application; After step S104, it further includes: S301. Obtain the type identifier of the test paper bag. The type identifier is used to group the test paper bags together, and the type identifier includes one or more of school, class, and printed product subject information; Among them, the type identifier of the test paper bag refers to a kind of marking information used to distinguish the attribution characteristics of different test paper bags, which refers to the relevant content that can reflect the unique attributes of the test paper bag and facilitate classification management, and is used to represent the classification situation of test paper bags in some key dimensions.

[0060] S302. Group the collected personnel activity trajectories through the type identifier; Specifically, on the basis of having obtained the type identifier of the test paper bag and collected the personnel activity trajectories, in order to deeply analyze the personnel operation conditions related to different types of test paper bags and judge whether the operations meet the specifications, etc., first sort out all the collected personnel activity trajectories, and then for each trajectory, check the type identifier corresponding to the test paper bag involved in the operation, and classify all the personnel activity trajectories according to the type identifier of the test paper bag, so that the subsequent analysis can focus on the personnel operation behaviors related to the same type of test paper bag and more accurately detect possible abnormal situations.

[0061] It should be noted that there are many types of data for the type identifier. However, in the actual operation process, in order to facilitate analysis and make the grouping more organized, one of the relevant data is selected here as the classification basis. The purpose is to make the activity trajectories comparable.

[0062] S303. Randomly select two collected personnel activity trajectories in the same group of collected personnel activity trajectories and judge the similarity; In some embodiments, for the position sequences of two trajectories, the numerical differentiation method is used to calculate the curvature value of each position point. After obtaining the curvature values of each position point of the two trajectories, the curvature values of their corresponding position points are compared point by point. Once it is found that the difference between the curvature values of a certain corresponding point is greater than the preset threshold, it means that there is a significant difference in the bending situation of the two trajectories at this position point. Adjust the similarity of the two trajectories, that is, reduce the similarity. The more the number of corresponding points with a curvature difference greater than the threshold, the more inconsistent the bending characteristics of the two trajectories at more positions, which also means that the behavior patterns of the collecting personnel at more operation positions are quite different. Correspondingly, the similarity of the two trajectories will be lower.

[0063] S304. If the similarity is lower than the similarity threshold, extract the avoidance behavior characteristics from the collected personnel activity trajectory with a later time sequence; Specifically, after calculating the similarity between two trajectories in the activity trajectories of the same group of collectors and comparing it with the preset similarity threshold, if it is found that the similarity is lower than the similarity threshold, it means that there are significant differences in the personnel operation behaviors reflected by these two trajectories. At this time, it is necessary to further analyze whether it is caused by situations such as avoiding test paper bags. The principle is that within the time range, it is possible that new test paper stacks are established or some existing test paper stacks are reduced. When such changes in the number and layout of test paper stacks occur, during the process of collectors performing test paper-related operation tasks, their activity trajectories will naturally change accordingly. First, focus on the activity trajectory of the collector with a later time sequence because if there is an avoidance behavior, it is often more likely to show corresponding trajectory change characteristics in subsequent operations. Then, carefully analyze each part of this trajectory to check whether there are situations such as a sudden change in the trajectory direction (a large-angle turn suddenly appears in a trajectory that was originally moving straight forward), a significant decrease in speed (suddenly decelerating in a section of the trajectory with normal speed), or a short pause (staying at a certain position for a time exceeding the normal operation stay time range). If these characteristics appear, extract them as avoidance behavior characteristics for subsequent use to find the corresponding obstacle test paper bags and further investigate abnormal operation situations.

[0064] In some specific embodiments, step S304 specifically includes: S3041. Obtain the position sequences and speed sequences of the two collector activity trajectories respectively; Specifically, perform image analysis on the surveillance video. According to the time sequence, extract the position information at different time points in the same trajectory and organize it into a position sequence; similarly, extract the corresponding speed information according to the time sequence to form a speed sequence.

[0065] S3042. Calculate the curvature values corresponding to the position sequences of the two collector activity trajectories; Specifically, for the position sequence on a plane, a curvature calculation formula based on the principle of calculus can be used. By analyzing and calculating the change in the tangent slope between adjacent coordinate points in the position sequence, the curvature value of each position point can be obtained.

[0066] S3043. In the activity trajectory of the collector with a later time sequence, when the curvature value is greater than the first preset threshold and the corresponding speed value is less than the second preset threshold, determine the candidate avoidance position; Specifically, when traversing in chronological order the position points corresponding to the activity trajectories of the collectors with later time sequences, for each position point, obtain its corresponding curvature value and speed value. Then, compare the curvature value of this position point with the first preset threshold, and at the same time compare the speed value with the second preset threshold. If it is found that the curvature value of this position point is greater than the first preset threshold and its speed value is less than the second preset threshold, this indicates that at this position, the trajectory of the collector has a large bend, and at the same time the movement speed significantly slows down. It is very likely that the collector encounters a situation that requires avoidance, such as a test paper bag blocking ahead, etc. Then, determine this position point as a candidate avoidance position, and screen out all position points that meet such conditions to form a candidate avoidance position set, preparing for further determination of whether there is a new avoidance behavior in the follow-up.

[0067] S3044. Determine whether the activity trajectory of the collector with an earlier time sequence at the corresponding spatial position satisfies that the curvature value is greater than the first preset threshold and the speed value is less than the second preset threshold; It should be noted that the principle and process of this step are similar to those of step S3043. The relevant principles and processes can refer to step S3043 and will not be limited here.

[0068] S3045. In the case of satisfaction, determine the candidate avoidance position as an avoidance behavior feature.

[0069] Specifically, after determining through the previous steps that the activity trajectory of the collector with an earlier time sequence at the corresponding spatial position also satisfies the condition that the curvature value is greater than the first preset threshold and the speed value is less than the second preset threshold, it indicates that these position points show similar characteristics of large bend and slow speed on both trajectories. It is very likely caused by a new avoidance behavior. Formally determine the previously determined candidate avoidance position as an avoidance behavior feature.

[0070] It can be seen that by obtaining the position and speed sequences of the activity trajectories of two collectors, the curvature value corresponding to the position sequence can reflect the bending degree of the trajectory, and the speed value reflects the movement speed. By setting thresholds, candidate avoidance positions with large curvature and small speed in the trajectory with a later time sequence are screened out, and then the corresponding spatial position conditions of the trajectory with an earlier time sequence are compared. The two confirm each other to determine the avoidance behavior feature.

[0071] S305. Identify the obstacle test paper bag corresponding to the avoidance behavior feature; obtain the first test paper bag corresponding to the activity trajectory of the collector with an earlier time sequence and the second test paper bag corresponding to the activity trajectory of the collector with a later time sequence; Specifically, based on the position information of the avoidance behavior characteristics in space, roughly determine the area range where obstacle test paper bags may exist. For example, if the avoidance behavior characteristics are manifested as a turn in a certain position trajectory, then take this turning point as the center and delimit a circular area with a certain radius as the search range. Then, within this search area, combine information such as the placement position and type identifier of the test paper bag to search for the corresponding test paper bag and determine it as an obstacle test paper bag. At the same time, obtain the first test paper bag corresponding to the activity trajectory of the collection personnel with an earlier time sequence and the second test paper bag corresponding to the activity trajectory of the collection personnel with a later time sequence. By checking the relevant information of these three test paper bags and their sequential relationship in the operation sequence, it is prepared for subsequent judgment of whether it conforms to the preset operating procedures.

[0072] In some embodiments, step S305 specifically includes: S3051. Construct a search area based on the avoidance behavior characteristics; Among them, the search area refers to a spatial area delimited based on the position where the avoidance behavior characteristics are located according to certain rules and ranges, and is used to search for relevant objects (here mainly obstacle test paper bags) that may cause avoidance behavior subsequently. It is equivalent to a scope limit for key investigation.

[0073] S3052. Determine the target area closest to the search area in the preset operating procedures; Specifically, consult the content such as the area division and corresponding function descriptions involved in each link of test paper bag storage and operation in the preset operating procedures, and sort out information such as the spatial position, range of different areas and their association with other areas. Then, by calculating the spatial distance or analyzing the area correlation, etc., compare the search area with each area mentioned in the preset operating procedures, and find the area that is closest to the search area or has the strongest correlation, and determine it as the target area.

[0074] S3053. Determine the test paper bag corresponding to the target area as the obstacle test paper bag.

[0075] It can be seen that based on the determined avoidance behavior characteristics, a search area is constructed, then the target area closest to it is determined according to the preset operating procedures, and then the corresponding test paper bag is locked as the obstacle test paper bag. Starting from the personnel avoidance behavior, it is accurately associated with the obstacle test paper bag that may affect the operation process.

[0076] S306. Judge according to the preset operating procedures whether the obstacle test paper bag meets the requirement of being carried after the first test paper bag and before the second test paper bag; Specifically, after the obstacle test paper bag has been identified and the first test paper bag and the second test paper bag have been determined, in order to deeply examine whether the operation process follows the preset sequence requirements, first, the specific regulations on the handling sequence of relevant test paper bags in the preset operation procedures are carefully checked. Then, the operation sequence relationship between the obstacle test paper bag and the first test paper bag and the second test paper bag in the actual situation is compared to check whether the obstacle test paper bag is indeed involved in the operation after the first test paper bag and before the second test paper bag. For example, if the preset operation procedures clearly require that when handling a batch of Chinese test papers, first handle the test paper bag on shelf A (i.e., the first test paper bag), then if there is a test paper bag with an obstacle situation that needs to be avoided at a specific position (i.e., the obstacle test paper bag), and then handle the test paper bag on shelf B (i.e., the second test paper bag), it is necessary to check whether the actual personnel movement trajectory and the test paper bag operation situation are consistent with it, so as to judge whether the operation sequence is compliant and avoid subsequent problems such as chaotic test paper distribution caused by incorrect sequence.

[0077] S307. If not all meet the requirements, the corresponding actions of the obstacle test paper bag, the first test paper bag, and the second test paper bag are determined as abnormal events.

[0078] It can be seen that determining the relative position relationship between the test paper bag areas can clarify the layout of each test paper bag stack in space. Then, the association strength of different test paper bag stacks is established, and the association strength decreases with the increase of the distance. In this way, the closeness of the association between each test paper bag stack can be measured according to the distance. Subsequently, the change amount of the distance between the test paper bags reduced or increased in the test paper bag stack and the belonging test paper bag stack is analyzed, and the group behavior consistency index is combined for comprehensive judgment. When there are increase or decrease changes in the test paper bag stack and meet the corresponding conditions, it is determined as the target test paper bag, and the overall relevance and change situation of the test paper bag stack are deeply considered, so as to distinguish whether it is a reasonable change in the normal handling and stacking process or a real batch chaos caused by improper handling, improving the accuracy of judging whether the test paper batch is abnormal.

[0079] Next, the material detection system 400 for an exemplary material stacking area provided by the embodiment of the present application is introduced. Figure 4 It is a schematic diagram of the exemplary hardware structure of the material detection system 400 for the material stacking area provided by the embodiment of the present application.

[0080] In some embodiments, the material detection system 400 in the material stacking area is a computer device or the material detection system 400 in the material stacking area includes a computer device. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with other external terminals or servers through a network connection. In some embodiments, the network interface can be a wired network interface, and in some embodiments, the network interface can also be a wireless network interface. The computer program, when executed by the processor, implements the method in the embodiments of the present application.

[0081] Those skilled in the art can understand that Figure 4 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component layout.

[0082] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

[0083] As used in the above embodiments, depending on the context, the term "when..." can be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" can be interpreted to mean "if determining..." or "in response to determining..." or "when detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)".

[0084] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive), etc.

[0085] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware with a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The foregoing storage medium includes: various media that can store program codes such as ROM or random access memory RAM, magnetic disks, or optical discs.

Claims

1. A method for detecting materials in a material stacking area, characterized in that, Including: Collecting surveillance images of the test paper storage area, and performing differential clarity processing on the surveillance images to obtain a sequence of feature maps, where the image clarity of the test paper bag area is higher than that of the non-test paper bag area. Within the test paper bag area, the clarity of the test paper bag contour, other areas, and the test paper bag body area decreases in sequence; increasing the resolution sampling for the simple area of the test paper bag contour and decreasing the resolution sampling for the complex area of the test paper bag contour; In the case of detecting a change in the position of the test paper bag, performing spatial registration calculation on the sequence of feature maps from before the previous position change to the current moment with a dynamic reference model to obtain the displacement amounts of different test paper bags; the dynamic reference model is a three-dimensional space mapping established for the initial stacking state when the test papers are stored in the warehouse; Determining the target test paper bags to be reduced or increased according to the displacement amounts of the test paper bags; Collecting the interaction data between the personnel activity trajectories and the test paper bag area, and performing matching analysis on the interaction data with a preset operation procedure; When the interaction data conforms to the preset operation procedure, determining the corresponding behavior as a normal event; When the interaction data does not conform to the preset work procedure, determining the corresponding behavior as an abnormal event; Judging the target test paper bags corresponding to the normal events with the preset work procedure; When the test paper bags for storage and retrieval specified in the preset work procedure do not correspond to the target test paper bags, changing the corresponding normal event to an abnormal event; Marking the abnormal events.

2. The method according to claim 1, wherein The step of determining the target test paper bags to be reduced or increased according to the displacement amounts of the test paper bags specifically includes: Determining the relative position relationship between the test paper bag areas; Establishing the association strength between different test paper bag stacks, where the association strength decays as the distance between the test paper bag stacks increases; the test paper bag stacks are all the test paper bags within the test paper bag area; Determining the change amount of the distance between the test paper bags to be reduced or increased in the test paper bag stack and the test paper bag stack to which they belong; Determining the group behavior consistency index according to all the change amounts; If the change amount of the distance between the test paper bags to be reduced or increased in the test paper bag stack and the test paper bag stack to which they belong is greater than a preset threshold and is outside the corresponding group behavior consistency index; determining them as target test paper bags.

3. The method according to claim 1, characterized in that, After the step of collecting the interaction data between the personnel activity trajectories and the test paper bag area and performing matching analysis on the interaction data with the preset operation procedure, the method further includes: Obtaining the type identifier of the test paper bag, which is used to group the test paper bags together, and the type identifier includes one or more of school, class, and printing product subject information; Grouping the personnel activity trajectories collected by the type identifier; Randomly selecting two personnel activity trajectories in the same group of the personnel activity trajectories collected and judging the similarity; If the similarity is lower than the similarity threshold, extracting the avoidance behavior characteristics from the personnel activity trajectory with a later time sequence; Identifying the obstacle test paper bags corresponding to the avoidance behavior characteristics; obtaining the first test paper bag corresponding to the personnel activity trajectory with an earlier time sequence and the second test paper bag corresponding to the personnel activity trajectory with a later time sequence; Judging whether the obstacle test paper bags conform to being carried after the first test paper bag and before the second test paper bag according to the preset operation procedure; If not all are met, the corresponding actions of the obstacle test paper bag, the first test paper bag, and the second test paper bag are determined as abnormal events.

4. The method according to claim 3, wherein The step of extracting the avoidance behavior feature from the activity track of the collector with a later extraction time sequence specifically includes: Obtain the position sequence and speed sequence of two collector activity tracks respectively; Calculate the curvature values corresponding to the position sequences of the two collector activity tracks; In the activity track of the collector with a later time sequence, when the curvature value is greater than the first preset threshold and the corresponding speed value is less than the second preset threshold, determine the candidate avoidance position; Judge whether the activity track of the collector with an earlier time sequence satisfies that the curvature value is greater than the first preset threshold and the speed value is less than the second preset threshold at the corresponding spatial position; In the case of satisfaction, determine the candidate avoidance position as the avoidance behavior feature.

5. The method according to claim 3, characterized in that, The step of identifying the obstacle test paper bag corresponding to the avoidance behavior feature specifically includes: Construct a search area in the avoidance behavior feature; Determine the target area closest to the search area in the preset operation procedure; Determine the test paper bag corresponding to the target area as the obstacle test paper bag.

6. The method according to claim 1, wherein The step of performing differential clarity processing on the monitoring image to obtain a sequence of feature maps specifically includes: Extract the boundary of the test paper bag in the monitoring image to obtain a preliminary test paper bag contour; Judge whether the preliminary test paper bag contour meets the requirements of a simple area; If it meets the requirements, construct a multi-scale feature pyramid, perform feature extraction on the test paper bag contour at different scales to obtain local fine features; Fuse the local fine features with the preliminary test paper bag contour to obtain the test paper bag contour structure.

7. The method according to claim 1, characterized in that After the step of marking the abnormal event, the method further includes: associating the corresponding batch identification information, and recording the operator's identity information, operation timestamp, and spatial position coordinates.

8. A material detection system for a material stacking area, characterized in that, The material detection system in the material stacking area includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the material detection system in the material stacking area to execute the method according to any one of claims 1-7.

9. A computer program product comprising instructions, characterized in that, When the computer program product runs on the material detection system in the material stacking area, it enables the material detection system in the material stacking area to execute the method according to any one of claims 1-7.

10. A computer-readable storage medium, comprising instructions, characterized in that, When the instruction runs on the material detection system in the material stacking area, it enables the material detection system in the material stacking area to execute the method according to any one of claims 1-7.

Images