Intelligent monitoring method and device for logistics sorting behavior
By combining scanning code information, indoor positioning data and high-definition camera images, violent sorting behavior is detected in real time and traced back to specific goods, the problem of difficulty in holding accountability in the existing technology is solved, and the efficiency and accuracy of logistics management are improved.
Patent Information
- Application Number
- CN202411179362.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-12-06
AI Technical Summary
The existing technology cannot accurately associate violent sorting with specific goods or waybills, resulting in difficulty in holding accountable and reducing logistics management efficiency.
By combining scanning code information, indoor positioning data and images captured by high-definition cameras, violent sorting behavior is detected in real time, its specific location and time is determined, and the stored information is traced back to the affected goods and waybills, using feature point matching and Similarity calculation ensures accuracy, combines video surveillance and real-time detection of sensor data, and optimizes camera configuration to improve detection accuracy.
It has achieved timely discovery and accurate accountability for violent sorting, improved the accuracy of goods handling and logistics management efficiency, and significantly reduced the false alarm rate and missed rate.
Smart Images

Figure CN119091376A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of logistics management, and in particular to an intelligent monitoring method for logistics sorting behavior. Background Art
[0002] With the rapid development of e-commerce, the scale of the logistics industry continues to expand. However, violent sorting has always been one of the problems that need to be solved in the logistics industry. Violent sorting not only damages goods, but may also lead to customer complaints and damage the company's reputation. Existing monitoring methods often cannot accurately associate violent sorting with specific goods or waybills, making it difficult to hold people accountable afterwards.
[0003] This solution aims to provide accurate association between violent sorting behavior and specific goods or waybills by combining scanned code information, indoor positioning data and images captured by high-definition cameras, thereby improving logistics management efficiency; As can be seen from the above, the problem of low logistics management efficiency still needs to be solved Summary of the invention
[0004] In order to improve the efficiency of logistics management, the present application provides an intelligent monitoring method for logistics sorting behavior.
[0005] In the first aspect, the present application provides an intelligent monitoring method for logistics sorting behavior, which adopts the following technical solutions:
[0006] An intelligent monitoring method for logistics sorting behavior, comprising: obtaining a timestamp and scan data of a code scanning action, and determining the location data of an operator, and determining the cargo data operated by the operator based on the timestamp and the location data, wherein the scan data includes corresponding transportation data and bill of lading data; determining a corresponding cargo image set based on the cargo data, wherein the image set includes a scanned cargo image and cargo images near the scanned cargo, and associating and storing the scan data, the cargo image and the timestamp; when violent sorting behavior is identified, determining a corresponding target image, determining a target time point data, determining a corresponding associated timestamp based on the time point, determining a corresponding cargo image set based on the associated timestamp, matching the target image with the cargo image set, determining a target cargo image, and retrieving target scan data corresponding to the target cargo image.
[0007] By adopting the above technical solution, violent sorting behavior can be detected in real time through video surveillance and sensor data to ensure timely discovery and action. Combined with the scan code timestamp and indoor positioning data, the specific location and time of the violent sorting behavior can be quickly determined, thereby locking the affected goods; then, through the associated stored information, the specific goods and waybills affected by the violent sorting behavior can be quickly traced, thereby improving the accuracy of cargo handling, strengthening monitoring and management, and significantly improving the efficiency of logistics management.
[0008] Optionally, in the process of matching the target image with the cargo image set to determine the target cargo image, the method also includes: determining a target feature point corresponding to the target image, retrieving cargo feature points corresponding to all cargo images in the cargo image set; calculating similarity between the target feature point and all cargo feature points, and obtaining corresponding similarity data; sorting all similarity data, and determining the target cargo based on the cargo feature point corresponding to the similarity data with the highest value.
[0009] By adopting the above technical solution, by extracting feature points from the target image and the cargo image set, key information in the image can be captured, and by matching the feature points of the target image with the feature points in the cargo image set, it is possible to ensure that the most similar image is found; by calculating the similarity between the target feature points and the cargo feature points, and sorting the similarity data, the cargo image corresponding to the similarity data with the highest value is selected as the target cargo image, it is possible to ensure that the target cargo is accurately found; through feature point matching and similarity calculation, a quick search can be achieved to determine the corresponding cargo image, thereby improving processing efficiency.
[0010] Optionally, in the process of determining the corresponding cargo image set based on the associated timestamp, the method also includes: calling a pre-set buffer time period, wherein the buffer time period is the working time of the operator for scanning the cargo code; determining the corresponding target time period based on the associated timestamp and the buffer time period; and determining the corresponding cargo image set based on the target time period.
[0011] By adopting the above technical solution, a buffer time period is set according to the working time of the operator scanning the goods, and the target time period is determined based on the associated timestamp and the buffer time period, thereby ensuring the rationality of the search range, more accurately locating images related to violent sorting behavior, reducing the interference of irrelevant images, thereby reducing the false alarm rate, and helping to improve the accuracy of finding the correct goods image.
[0012] Optionally, in the process of determining the corresponding set of cargo images based on the target time period, the method further includes: determining the corresponding target area after identifying the violent sorting behavior; retrieving the corresponding target time period; and determining the corresponding set of cargo images based on the target area and the target time period.
[0013] By adopting the above technical solution, the spatial scope of the search can be limited by determining the target area, and the time range of the search can be limited by combining the determined target time period. By limiting the spatial and time range of the search, the interference of irrelevant images can be reduced, thereby reducing the false alarm rate and improving the efficiency of image search.
[0014] Optionally, when identifying violent sorting behavior, the method also includes: obtaining video streams from multiple cameras in the sorting area; decoding based on the video stream to obtain a corresponding image sequence; calling a pre-trained sorting recognition model, and inputting the image sequence into the sorting recognition model for detection; if the feature data matches successfully, it is determined that violent sorting behavior exists.
[0015] By adopting the above technical solution, by decoding based on video streams and using pre-trained sorting recognition models for detection to identify violent sorting behavior, the system's real-time monitoring capability, recognition accuracy and overall performance can be significantly improved. It can not only improve the response speed of the monitoring system, but also reduce false alarms and missed alarms, thereby effectively improving logistics management efficiency.
[0016] Optionally, the method further includes: acquiring environmental data of the sorting area, wherein the environmental data includes light intensity, temperature and humidity, and brightness of the sorting area; dividing the sorting area into shooting areas based on the environmental data; and adjusting the camera parameters in the shooting area for each shooting area based on the environmental data.
[0017] By adopting the above technical solution, by obtaining the environmental data of the sorting area and adjusting the camera parameters based on the environmental data, the image quality can be significantly improved, the camera configuration can be optimized, and the accuracy of detecting violent sorting behaviors can be improved. This can not only improve the performance of the monitoring system, but also ensure that violent sorting behaviors can be accurately captured under various environmental conditions, thereby improving the overall efficiency of logistics management.
[0018] In the second aspect, the present application provides an intelligent monitoring device for logistics sorting behavior, which adopts the following technical solution:
[0019] An intelligent monitoring device for logistics sorting behavior, comprising:
[0020] A cargo data determination module, which obtains the timestamp and the scanned data of the code scanning action, and determines the location data of the operator, and is used to determine the cargo data operated by the operator based on the timestamp and the location data, wherein the scanned data includes the corresponding transportation data and cargo list data;
[0021] an association module, which determines a corresponding set of cargo images based on the cargo data, wherein the set of images includes scanned cargo images and cargo images near the scanned cargo, and is used to associate and store the scanned cargo data, the cargo images, and the timestamp;
[0022] The target scan code data determination module determines the corresponding target image and the target time point data when violent sorting behavior is identified, determines the corresponding associated timestamp based on the time point, determines the corresponding cargo image set based on the associated timestamp, matches the target image with the cargo image set, determines the target cargo image, and is used to determine the target scan code data corresponding to the target cargo image.
[0023] In the third aspect, the present application provides an intelligent monitoring method for logistics sorting behavior, which adopts the following technical solution: an intelligent monitoring method for logistics sorting behavior, comprising a processor, in which a program of the intelligent monitoring method for logistics sorting behavior described in any one of the above items is running.
[0024] In a fourth aspect, the present application provides a storage medium, which adopts the following technical solution:
[0025] A storage medium storing a program of the intelligent monitoring method for logistics sorting behavior described in any one of the above.
[0026] In summary, the present application includes at least one of the following beneficial technical effects: real-time detection of violent sorting behavior through video surveillance and sensor data to ensure timely discovery and action; combined with the code scanning timestamp and indoor positioning data, the specific location and time of the violent sorting behavior can be quickly determined, thereby locking the affected goods; then, through the associated stored information, the specific goods and waybills affected by the violent sorting behavior can be quickly traced back, thereby improving the accuracy of cargo handling, strengthening monitoring and management, and significantly improving the efficiency of logistics management. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 is a flow chart of an intelligent monitoring method for logistics sorting behavior according to an exemplary embodiment;
[0028] Figure 2 It is a structural block diagram of an intelligent monitoring device for logistics sorting behavior according to an exemplary embodiment. DETAILED DESCRIPTION
[0029] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.
[0030] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0031] The present application embodiment discloses an intelligent monitoring method for logistics sorting behavior, referring to Figure 1 ,include:
[0032] S100, obtaining the timestamp and data of the code scanning action, and determining the location data of the operator, and based on the timestamp and the location data, determining the cargo data on which the operator performs the operation.
[0033] Among them, the scanning data includes the corresponding transportation data and cargo list data; in addition, when the operator uses a barcode scanner to scan the goods, the system records the exact timestamp and scanning data of the scanning action, and then tracks the operator's location in real time through the indoor positioning system. Combined with the scanning timestamp and the operator's location information, the system can determine the goods the operator is operating and the sorting area where the goods are located.
[0034] S110, determining a corresponding set of cargo images based on the cargo data, and storing the scanned code data, cargo images, and timestamps in association with each other.
[0035] Among them, the image set includes the image of the scanned goods and the image of the goods near the scanned goods. The high-definition camera installed in the sorting area can be used to capture the scanning action and obtain the corresponding goods image, so as to obtain the corresponding goods image set. Then, the scanning data (including transportation data, bill of lading data), goods image and timestamp information can be associated and stored, so as to ensure the accuracy of each operation link.
[0036] S120, when violent sorting behavior is identified, determine the corresponding target image, determine the target time point data, determine the corresponding associated timestamp based on the time point, determine the corresponding cargo image set based on the associated timestamp, match the target image with the cargo image set, determine the target cargo image, and retrieve the target scan code data corresponding to the target cargo image.
[0037] First, violent sorting behavior can be detected in real time through video surveillance and sensor data. At this time, the target image corresponding to the goods where the violent sorting behavior occurs can be obtained through the camera, and the target time point data when the violent sorting behavior occurs can be recorded, and the corresponding associated timestamp can be determined according to the time point. The corresponding goods image set is determined by the associated timestamp. At this time, it is equivalent to dividing a smaller image set from all the collected goods images, and then matching the target image with the goods image set. After the match is successful, the successfully matched goods image is determined as the target goods image, and because the goods image is associated with the corresponding scan code data, the target scan code data corresponding to the target goods image can be retrieved.
[0038] Through the above process, combined with the timestamp of the code scan and the indoor positioning data, the specific location and time of the violent sorting behavior can be quickly determined, thereby locking the affected goods. By associating the stored information, the specific goods and waybills affected by the violent sorting behavior can be quickly traced back, reducing the investigation time. The image analysis technology is applied in the middle to automatically identify the image of the goods, reducing the workload of the monitoring personnel.
[0039] It should be noted that, in the process of matching the target image with the cargo image set to determine the target cargo image, the method further includes:
[0040] S121, determining the target feature points corresponding to the target image, and retrieving the cargo feature points corresponding to all cargo images in the cargo image set.
[0041] Among them, feature extraction algorithms (such as SIFT, SURF, etc.) are used to extract key feature points from the target image. By extracting feature points in the target image and the cargo image set, the key information in the image can be captured, thereby improving the matching accuracy.
[0042] S122, calculating the similarity between the target feature point and all the cargo feature points, and obtaining corresponding similarity data.
[0043] Among them, by calculating the similarity between the target feature points and the cargo feature points, it is possible to quickly determine which cargo images are most similar to the target image.
[0044] S123, sorting all the similarity data, and determining the target goods based on the goods feature points corresponding to the similarity data with the highest value.
[0045] Among them, the similarity data are sorted, and the cargo image corresponding to the similarity data with the highest value is selected as the target cargo image, which can ensure that the target cargo is found accurately.
[0046] Here, after calculating the similarity data, by setting a suitable similarity threshold, the similarity data is compared with the similarity threshold, and the similarity data less than the similarity threshold is filtered out, thereby filtering out unmatched images and reducing false alarms, thereby realizing a multi-level screening mechanism, first performing a rough screening, and then performing a fine match, to further reduce the false alarm rate.
[0047] By executing steps S121 to S123, the accuracy and efficiency of the system in determining the target cargo image can be improved, so that the affected cargo can be quickly and accurately located when dealing with violent sorting behavior, which can improve the overall performance of the system and facilitate logistics management.
[0048] In addition, in the process of determining the corresponding cargo image set, the following steps are also performed:
[0049] S124, calling a pre-set buffer time period.
[0050] The buffer time period is pre-set according to the working time of the operator in scanning the goods. For example, if the operator needs an average of 10 seconds to complete a scanning action, the buffer time period can be set to 10 seconds before and after. In addition, the buffer time period can be adjusted according to actual conditions to adapt to different operating speeds and duration of violent sorting time.
[0051] S125, determining a corresponding target time period based on the associated timestamp and the buffer time period;
[0052] Among them, if the violent sorting behavior occurs at 10:00:30, and the buffer period is 10 seconds before and after, then the target time period is 10:00:20 to 10:00:40.
[0053] S126, after identifying the violent sorting behavior, the system can determine the corresponding target area.
[0054] Among them, by determining the specific area where the violent sorting incident occurred, the search scope can be narrowed and the accuracy of finding relevant images can be improved, which can then be combined with determining the target time period to further narrow the scope of the cargo image set.
[0055] S127, determining a corresponding cargo image set based on the target area and the target time period
[0056] That is, during the time period from 10:00:20 to 10:00:40, all cargo images collected in the sorting area corresponding to the target area are retrieved. By executing the above steps S124 to S127, the accuracy and efficiency of the system in determining the cargo image set can be significantly improved, which can not only reduce the search range and improve the search efficiency, but also reduce the false alarm rate.
[0057] In the embodiment of the present application, it is necessary to identify violent sorting behavior. Here, a pre-trained sorting recognition model is used for recognition. The training process of the sorting recognition model is as follows:
[0058] To select a suitable deep learning model based on the needs of sorting and recognition, the labeled video data needs to be input into the deep learning model to train the model. The deep learning model will learn how to extract features from the video and match these features with the violent sorting behavior pattern; then, the training effect and generalization ability of the deep learning model can be improved by adjusting the deep learning model parameters and optimization algorithms. Among them, the deep learning model parameters generally refer to the learning rate, batch size, number of iterations, etc., and the optimization algorithm generally refers to the gradient descent method, Adam optimizer, etc. Finally, the deep learning model is evaluated using the validation set, and the deep learning model structure and parameters are adjusted according to the evaluation results until the deep learning model achieves satisfactory performance, and the deep learning model is determined as the sorting and recognition model to be used later.
[0059] The specific process of identifying violent sorting behavior is as follows:
[0060] S1201, obtaining video streams from multiple cameras in a sorting area, decoding based on the video streams, and obtaining corresponding image sequences.
[0061] The video stream obtained from the camera is decoded into a series of image frames, and the decoded images are combined into an image sequence in chronological order.
[0062] S1202, retrieve a pre-trained sorting and recognition model, and input the image sequence into the sorting and recognition model for detection.
[0063] Among them, a pre-trained sorting and recognition model is called, and the sorting and recognition model extracts features from the input image sequence. The sorting and recognition model determines whether there is violent sorting behavior based on the extracted features.
[0064] S1203: If the feature data matches successfully, it is determined that violent sorting behavior exists.
[0065] The extracted features are matched with the features of the violent sorting behavior, and if the match is successful, it is determined that the violent sorting behavior exists.
[0066] By training the model to identify the characteristics of violent sorting behavior, violent sorting behavior is determined only when the characteristic data is matched successfully, which can reduce false alarms and thus help reduce the false alarm rate.
[0067] Finally, it should be pointed out that in the process of monitoring the logistics sorting area, the environment of the logistics sorting area will change with different time periods. It is necessary to adjust the parameters of the high-definition camera in time so that a relatively high-quality video stream can be obtained throughout the process. Therefore, the method also includes:
[0068] S105, obtaining environmental data of the sorting area.
[0069] Among them, environmental data includes light intensity, temperature, humidity, and brightness of the sorting area.
[0070] S106, dividing the sorting area into shooting areas based on the environmental data.
[0071] Among them, according to different environmental data, the sorting area is divided into several shooting area sets, and a targeted camera configuration strategy is formulated for each shooting area set.
[0072] S107, adjusting the camera parameters in each shooting area based on the environmental data.
[0073] For each shooting area, the camera parameters such as exposure time, gain, white balance, etc. are automatically adjusted according to the environmental data of the area (such as light intensity, temperature, humidity and brightness, etc.), and the optimized configuration ensures that the camera in each shooting area can capture high-quality images.
[0074] By obtaining environmental data of the sorting area and adjusting camera parameters based on the environmental data, it is possible to ensure that high-quality video streams are captured in different environments. By optimizing the camera configuration, it is not only possible to ensure that violent sorting behaviors can be accurately captured under various environmental conditions, but also to improve the overall efficiency of logistics management.
[0075] The present application embodiment discloses an intelligent monitoring device for logistics sorting behavior, referring to Figure 2 ,include:
[0076] The cargo data determination module 200 obtains the timestamp and the scanned data of the code scanning action, and determines the location data of the operator, and is used to determine the cargo data operated by the operator based on the timestamp and the location data, wherein the scanned data includes the corresponding transportation data and cargo list data;
[0077] An association module 210 determines a corresponding cargo image set based on the cargo data, wherein the image set includes a scanned cargo image and cargo images near the scanned cargo, and is used to associate and store the scanned cargo data, cargo image, and timestamp;
[0078] The target scan code data determination module 220, when identifying violent sorting behavior, determines the corresponding target image, determines the target time point data, determines the corresponding associated timestamp based on the time point, determines the corresponding cargo image set based on the associated timestamp, matches the target image with the cargo image set, determines the target cargo image, and is used to determine the target scan code data corresponding to the target cargo image.
[0079] Optionally, the device includes but is not limited to:
[0080] A target feature point determination module is used to determine the target feature points corresponding to the target image, and retrieve the cargo feature points corresponding to all cargo images in the cargo image set;
[0081] A similarity calculation module is used to calculate the similarity between the target feature points and all the cargo feature points, and obtain the corresponding similarity data;
[0082] The target goods determination module sorts all the similarity data and uses the goods feature points corresponding to the similarity data with the highest value to determine the target goods.
[0083] Optionally, the device includes but is not limited to:
[0084] A buffer time period calling module is used to call a preset buffer time period, wherein the buffer time period is the working time of the operator for scanning the goods;
[0085] A target time period determination module, for determining a corresponding target time period based on an associated timestamp and a buffer time period;
[0086] The cargo image set determination module is used to determine the corresponding cargo image set based on the target time period.
[0087] Optionally, the device includes but is not limited to:
[0088] The target area determination module determines the corresponding target area after identifying the violent sorting behavior;
[0089] A target time period calling module is used to call the corresponding target time period;
[0090] The cargo image set determination module is used to determine the corresponding cargo image set based on the target area and the target time period.
[0091] Optionally, the device includes but is not limited to:
[0092] A video stream acquisition module acquires video streams from multiple cameras in a sorting area, wherein the resolution of a low-resolution stream is lower than that of a high-resolution stream;
[0093] An image sequence acquisition module is used to acquire a corresponding image sequence based on decoding of the video stream;
[0094] The detection module retrieves the pre-trained sorting and recognition model and inputs the image sequence into the sorting and recognition model for detection;
[0095] If the feature data matches successfully, it is determined that violent sorting behavior has occurred.
[0096] Optionally, the device includes but is not limited to:
[0097] An environmental data acquisition module is used to acquire environmental data of the sorting area, wherein the environmental data includes light intensity, temperature, humidity, and brightness of the sorting area;
[0098] A shooting area division module, used to divide the sorting area into shooting areas based on environmental data;
[0099] The camera parameter adjustment module is used to adjust the camera parameters in each shooting area based on the environmental data.
[0100] An embodiment of the present application further discloses an intelligent monitoring method for logistics sorting behavior, including a processor, in which a program of any one of the above-mentioned intelligent monitoring methods for logistics sorting behavior is running.
[0101] The embodiment of the present application also discloses a storage medium storing a program of any one of the above-mentioned intelligent monitoring methods for logistics sorting behavior.
[0102] In summary, the buffer time period is set according to the working time of the operator scanning the goods code, the target time period is determined based on the associated timestamp and the buffer time period, and combined with the target area corresponding to the violent sorting behavior, the rationality of the search range is ensured, and the images related to the violent sorting behavior can be located more accurately, and the interference of irrelevant images can be reduced, thereby reducing the false alarm rate and helping to improve the accuracy of finding the correct goods image; in addition, using the pre-trained sorting recognition model for detection to identify violent sorting behavior can significantly improve the real-time monitoring capability, recognition accuracy and overall performance of the system, which can not only improve the response speed of the monitoring system, but also effectively improve the efficiency of logistics management; optimizing the camera configuration and improving the accuracy of violent sorting behavior detection can not only improve the performance of the monitoring system, but also ensure that violent sorting behavior can be accurately captured under various environmental conditions, thereby improving the overall efficiency of logistics management.
[0103] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. An intelligent monitoring method for logistics sorting behavior, It is characterized in that include: Obtaining a timestamp and scan data of the code scanning action, and determining the location data of the operator, and determining the cargo data operated by the operator based on the timestamp and the location data, wherein the scan data includes corresponding transportation data and cargo list data; Determine a corresponding cargo image set based on the cargo data, wherein the image set includes a scanned cargo image and a cargo image near the scanned cargo, and store the scanned cargo data, the cargo image, and the timestamp in association; When violent sorting behavior is identified, the corresponding target image is determined, the target time point data is determined, the corresponding associated timestamp is determined based on the time point, the corresponding cargo image set is determined based on the associated timestamp, the target image is matched with the cargo image set, the target cargo image is determined, and the target scan code data corresponding to the target cargo image is retrieved.
2. According to claim 1, the intelligent monitoring method for logistics sorting behavior, It is characterized in that In the process of matching the target image with the set of cargo images to determine the target cargo image, the method further includes: Determine the target feature points corresponding to the target image, and retrieve the cargo feature points corresponding to all cargo images in the cargo image set; Calculating the similarity between the target feature point and all the cargo feature points, and obtaining corresponding similarity data; All similarity data are sorted, and the target goods are determined based on the goods feature points corresponding to the similarity data with the highest value.
3. According to claim 1, the intelligent monitoring method for logistics sorting behavior, It is characterized in that In the process of determining the corresponding cargo image set based on the associated timestamp, the method further includes: Retrieve the pre-set buffer time period, where the buffer time period is the working time of the operator for scanning the goods; Determining a corresponding target time period based on the associated timestamp and the buffer time period; A corresponding cargo image set is determined based on the target time period.
4. According to claim 1, the intelligent monitoring method for logistics sorting behavior, It is characterized in that In the process of determining the corresponding cargo image set based on the target time period, the method further includes: After identifying violent sorting behavior, determine the corresponding target area; Retrieve the corresponding target time period; A corresponding cargo image set is determined based on the target area and the target time period.
5. According to claim 1, the intelligent monitoring method for logistics sorting behavior, It is characterized in that When identifying violent sorting behavior, the method also includes: Get video streams from multiple cameras in the sorting area; Decoding the video stream to obtain a corresponding image sequence; Retrieving a pre-trained sorting and recognition model, and inputting the image sequence into the sorting and recognition model for detection; If the characteristic data matches successfully, it is determined that violent sorting behavior occurs.
6. The intelligent monitoring method for logistics sorting behavior according to claim 1, It is characterized in that The method further comprises: Obtaining environmental data of the sorting area, wherein the environmental data includes light intensity, temperature, humidity, and brightness of the sorting area; Dividing the sorting area into shooting areas based on the environmental data; For each shooting area, the camera parameters in the shooting area are adjusted based on the environmental data.
7. An intelligent monitoring device for logistics sorting behavior, It is characterized in that include: A cargo data determination module, which obtains the timestamp and the scanned data of the code scanning action, and determines the location data of the operator, and is used to determine the cargo data operated by the operator based on the timestamp and the location data, wherein the scanned data includes the corresponding transportation data and cargo list data; an association module, which determines a corresponding set of cargo images based on the cargo data, wherein the set of images includes scanned cargo images and cargo images near the scanned cargo, and is used to associate and store the scanned cargo data, the cargo images, and the timestamp; The target scan code data determination module determines the corresponding target image and the target time point data when violent sorting behavior is identified, determines the corresponding associated timestamp based on the time point, determines the corresponding cargo image set based on the associated timestamp, matches the target image with the cargo image set, determines the target cargo image, and is used to determine the target scan code data corresponding to the target cargo image.
8. An intelligent monitoring method for logistics sorting behavior, It is characterized in that It comprises a processor, in which runs a program of the intelligent monitoring method for logistics sorting behavior as described in any one of claims 1 to 6.
9. A storage medium, It is characterized in that A program storing the intelligent monitoring method of logistics sorting behavior as described in any one of claims 1 to 6.
Citation Information
Patent Citations
Video analysis-based sorter operation specification detection method
CN108230349A
Violation sorting tracing method and device and storage medium
CN111489123A
Intelligent logistics sorting method based on Internet of Things
CN112474375A
Logistics data security processing method and system, electronic equipment and computer storage medium
CN112668966A
