Visual logistics tracing system and method based on video surveillance network

Abstract

The present invention discloses a visual logistics tracing system and method based on a video surveillance network, which relates to the field of logistics tracing technology. The communication coefficient is constructed by the acquired communication quality data, and the flow process of express items in the storage area is monitored by the installed recognition module to obtain the image data of the express items during the flow; the target items are screened out by the deformation degree of the express items at each logistics node, and the corresponding loss density is obtained by analyzing the target item data at the logistics node. If the loss density is higher than the density threshold, the logistics information chain of the target item is constructed by the acquired image data, and the number of damaged express items in the sub-area is determined by the logistics information chain and the damage coefficient of the sub-area is constructed. If it exceeds expectations, the corresponding optimization plan is given by the logistics warehousing optimization knowledge map. The traceability processing of the target item is realized, the position and corresponding area where the target item is deformed and damaged are determined, and the optimization target is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics tracing, and in particular to a visual logistics tracing system and method based on a video monitoring network. Background Art

[0002] Logistics warehousing management refers to the process of effectively managing logistics warehouses and inventories through scientific, standardized and efficient methods. It involves the flow of logistics information, the process of logistics operations, the operation of logistics equipment, the coordination of logistics personnel, etc., aiming to improve logistics operation efficiency, reduce logistics costs, and improve customer satisfaction. For example: inventory management, warehouse layout and design, operation process optimization and logistics information management,

[0003] In a Chinese invention patent with application publication number CN116629720A, a logistics traceability method is disclosed, including the following steps: establishing an enterprise registration account, associating multiple enterprises for the enterprise registrant account, and associating each enterprise with an employee; creating a sales website or an online store on an online sales platform; applying for a product number and a product QR code for each product; then transmitting the product number and the product QR code to the sales website or transmitting the product number and the product QR code to the online store on the online sales platform; the present invention uses an intelligent scanning instrument to scan the express packaging box, measures the length, width and height of the express packaging box, and then transmits the information to the sales website or sales platform through the network, so that it can be known whether the express has been opened by someone during transportation, so that the interests of the buyer will not be harmed.

[0004] Combined with the above application and the contents of the prior art:

[0005] After the goods are transported to the storage area, they will enter the transportation and distribution stage. During this stage, a series of operations such as information entry and classification of the goods need to be performed, and after the next destination of each goods is determined, the goods will be sent out in time; however, after the goods enter the storage area, the amount of goods is usually large and the processing time is relatively tight. This leads to the goods, that is, express items, being damaged or deformed during circulation in the storage area due to various improper operations. Such damage and deformation may not even be discovered until they are delivered to the user. Since it is not easy to determine the cause of such damage in the storage center, it is difficult to clearly divide the responsibility for damage;

[0006] In order to solve the above problems, the existing logistics traceability method usually attaches a QR code to the outside of the express item. By identifying the QR code, the various areas and paths that the goods have passed through are determined. When damage occurs, the link where the damage occurred is inferred. Although this express item traceability method can determine the circulation process of the express item, when it is damaged, the corresponding traceability result is not clear enough, and it is difficult to effectively determine the specific link of the damage, and the traceability effect is poor.

[0007] To this end, the present invention provides a visual logistics tracing system and method based on a video monitoring network. Summary of the invention

[0008] 1. Technical issues to be resolved

[0009] In view of the deficiencies of the prior art, the present invention provides a visual logistics tracing system and method based on a video surveillance network. By acquiring image data of express items during circulation, the target items are screened out by the degree of deformation of the express items at each logistics node, and the corresponding loss density is obtained by analyzing the target item data at the logistics node. If the loss density is higher than the density threshold, the logistics information chain of the target item is constructed by the acquired image data, and the number of damaged express items in the sub-region is determined by the logistics information chain and the damage coefficient of the sub-region is constructed. If it exceeds expectations, the corresponding optimization plan is given by the logistics warehousing optimization knowledge map. The traceability of the target item is realized, the location and corresponding area where the target item is deformed and damaged are determined, and the optimization target is improved, thereby solving the technical problems raised in the background technology.

[0010] (II) Technical solution

[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0012] The visual logistics tracing method based on the video surveillance network includes collecting express loss data in the storage area, constructing the loss degree Sts in the storage area based on the express loss data, and if the loss degree Sts exceeds expectations, screening out risk nodes from several logistics nodes in the storage area based on the amount of express passing through each logistics node;

[0013] A monitoring module is added at the risk node, and the communication quality data in the vicinity of the logistics node is detected. The communication coefficient Xts is constructed based on the acquired communication quality data. If the communication coefficient Xts is lower than the quality threshold, the data transmission mode of the monitoring module is switched.

[0014] The installed identification module monitors the flow of express items in the storage area, and uses the acquired monitoring data and related data to train and build a digital twin model of express item flow, and uses real-time data to update the model;

[0015] Obtain image data of express parcels during circulation, and select target parcels based on the degree of deformation at each logistics node through image recognition and comparison. Analyze the target parcel data at the logistics node to obtain the corresponding loss density Mj. If the loss density Mj is higher than the density threshold, issue a damaged parcel reminder.

[0016] The logistics information chain of the target item is constructed from the acquired image data. The damaged quantity of express items in the sub-area is determined by the logistics information chain and the damage coefficient Qts of the sub-area is constructed. If the damage coefficient Qts exceeds expectations, the logistics warehousing optimization knowledge graph gives a corresponding optimization plan; the damage coefficient Qts is obtained as follows: linearly normalize the damaged quantity Xq, and map the corresponding data value to the interval [0, 1], and then follow the following formula:

[0017]

[0018] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=1, 2, …, p, where p is the number of sub-periods; Xq i is the number of damaged express items in the i-th sub-period, Xq avg is the mean number of damaged items, is the qualified standard value for the number of damaged items.

[0019] Furthermore, the storage area is used as the monitoring area, and the loss amount Su of express items in the monitoring area is counted in each sub-period, and the loss amount data in each sub-area is aggregated to construct a data set of express item loss; the loss degree Sts is constructed from the express item loss data set, and if the loss degree Sts exceeds the preset loss threshold, a screening instruction is issued to the outside;

[0020] The loss degree Sts is constructed as follows: After dimensionless processing of the loss quantity Su, it is constructed as follows:

[0021]

[0022] Weight coefficient: 0≤F 1 ≤1,0≤F 2 ≤1 and F 2 +F 1 =1; i = 1, 2, ..., p, p is the number of sub-periods, Su i is the loss amount of express delivery in the ith sub-period, Su avg is the mean of the loss amount; The acceptable value of loss is the loss amount.

[0023] Furthermore, after receiving the screening instruction, the logistics nodes are marked in the monitoring area, the logistics nodes are marked on the electronic map, and the number of express parcels passing through each logistics node Tp is counted in the sub-period; the risk level Fts is constructed based on the number of express parcels. If the risk level Fts exceeds the preset risk threshold, the corresponding logistics node is used as a risk node, and a supplementary instruction is issued to the outside.

[0024] The risk level Fts is constructed as follows. After dimensionless processing of the express delivery volume Tp, it is constructed as follows:

[0025]

[0026] Among them, F 1 、F 2 is the weight coefficient, 0≤F 1 ≤1,0≤F 2 ≤1, and F 1 +F 2 =1, i=1, 2, ... p, p is the number of sub-periods, Tpi is the number of express parcels at the logistics node in the i-th sub-period, Tpau g is the mean number of express parcels.

[0027] Furthermore, after receiving the supplementary instruction, a monitoring module is added at the risk node, and the installation position of the monitoring module is constrained in the following manner:

[0028]

[0029] Where n is the number of monitoring modules, V(i, j) is the distance from monitoring module i to monitoring module j, and S avg is the mean distance, Fts is the risk of the logistics node, and Tp is the number of express parcels passing through the logistics node.

[0030] Furthermore, the communication quality in the vicinity of the logistics node is monitored, and the acquired communication quality data is aggregated to construct a communication status data set; the communication coefficient Xts is generated from the communication status data set, where the bit error rate Ym and the communication efficiency Tm are linearly normalized, and the corresponding data values ​​are mapped to the interval [0, 1], and then according to the following method:

[0031]

[0032] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1; if the communication coefficient Xts is lower than the preset quality threshold, different data transmission modes are switched when the signal strength changes.

[0033] Furthermore, when express items flow between various logistics nodes, identification modules are installed on the logistics nodes and corresponding transportation vehicles to record and track logistics information and obtain the location information of each express item. The identification module collects the location data and corresponding time data of the express items, and collects the equipment and related data when distributing and transferring the express items, and then summarizes them to construct a collection of express item transportation flow data.

[0034] Furthermore, the monitoring module obtains image data of the express items during circulation at the logistics nodes, and the image recognition model performs object detection and recognition on the obtained image data, identifies the relevant data of the express items and the means of transport, and determines the corresponding location information after identifying the express items; compares the current image data of the express items with the initial image data, determines the degree of deformation of the express items at each logistics node, and uses the degree of deformation as the deformation ratio; if the deformation ratio exceeds expectations, the corresponding express item will be determined as a target item.

[0035] Furthermore, the target items are marked in the digital twin model of express delivery flow. After obtaining the target item data at each logistics node, the corresponding loss density Mj is analyzed under dimensionless conditions as follows:

[0036]

[0037] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=1, 2, …k, k is the number of target parts; D ij is the shortest distance from the i-th target part to the j-th target part, Df avg is the distance mean; W ij is the difference in flow speed between the i-th target part and the j-th target part, Wfav g is the mean value of the flow velocity difference.

[0038] Furthermore, after receiving the damaged part reminder, the target part information is obtained, and the independent frame containing the target part information is extracted from the monitoring video. Several independent frames are connected according to the acquisition time to form the logistics information chain of the target part; the monitoring area is divided into several sub-areas, and after the damaged position of the target part is determined by the logistics information chain, the number of damaged express items in the sub-area in each sub-period is determined, and the damage coefficient Qts of the sub-area is constructed. If the damage coefficient Qts exceeds expectations, the corresponding sub-area is regarded as an abnormal area. If the proportion of abnormal areas exceeds the expected proportion, an optimization instruction is issued to the outside;

[0039] After receiving the optimization instruction, the corresponding optimization features are identified and obtained, and the logistics warehousing optimization is used as the target word to pre-construct the logistics warehousing optimization knowledge graph; based on the correspondence between the optimization features and the optimization plan, the logistics warehousing optimization knowledge graph gives the corresponding optimization plan, and the optimization plan is used to optimize the express delivery flow process in the monitoring area.

[0040] The visual logistics traceability system based on the video surveillance network includes: a risk analysis unit, which collects express loss data in the storage area, and constructs the loss degree Sts in the storage area based on the express loss data. If the loss degree Sts exceeds expectations, the risk nodes are screened out from several logistics nodes in the storage area according to the amount of express delivery passing through each logistics node;

[0041] The communication quality detection unit adds a monitoring module at the risk node and detects the communication quality data in the vicinity of the logistics node. The communication coefficient Xts is constructed based on the acquired communication quality data. If the communication coefficient Xts is lower than the quality threshold, the data transmission mode of the monitoring module is switched.

[0042] The model building unit uses the installed recognition module to monitor the flow of express items in the storage area, and uses the acquired monitoring data and related data to train and build a digital twin model of the express item flow, and uses real-time data to update the model;

[0043] The loss monitoring unit obtains image data of express items during circulation, and selects target items based on the degree of deformation of express items at each logistics node through image recognition and comparison. The corresponding loss density Mj is obtained by analyzing the target item data at the logistics node. If the loss density Mj is higher than the density threshold, a damaged item reminder is issued.

[0044] The logistics optimization unit constructs the logistics information chain of the target item based on the acquired image data. The logistics information chain determines the number of damaged express items in the sub-area and constructs the damage coefficient Qts of the sub-area. If the damage coefficient Qts exceeds expectations, the logistics warehousing optimization knowledge graph gives a corresponding optimization plan.

[0045] (III) Beneficial effects

[0046] The present invention provides a visual logistics tracing system and method based on a video surveillance network, which has the following beneficial effects:

[0047] 1. Determine the corresponding logistics nodes and calculate the risk level Fts at each logistics node. Select the risk nodes from several logistics nodes based on the risk level Fts. This will help achieve more obvious optimization goals when optimizing the transportation and distribution of express items. It will also help improve processing efficiency and targeted processing when actually processing.

[0048] 2. By increasing the number of monitoring modules, when collecting images of express items, the amount of image data collected can be increased, thereby improving the accuracy of tracking when tracking express items.

[0049] 3. Construct the corresponding communication coefficient Xts from the acquired detection data, and evaluate and judge the communication capacity at the logistics node based on the communication coefficient Xts. If the communication capacity is good, the current data transmission mode can be maintained. If the current communication quality is poor, switch the current data transmission mode to ensure and adjust the current communication quality, so that the data can be sent out in time after the image data or other data is acquired.

[0050] 4. Train and obtain the digital twin model of express delivery flow. When managing express delivery in the storage area, after obtaining the location and image data of the express delivery, the digital twin model of express delivery flow is constructed to display the express delivery process. It can also predict the flow and distribution of express delivery in the storage area.

[0051] 5. Through image recognition and comparison, each express item is screened according to its deformation condition, and those with higher deformation degree are selected as target items. By screening out the target items, they can be processed in time to reduce the losses caused by product deformation or damage.

[0052] 6. Judge the loss degree of express items in the current storage area based on the loss density Mj. If the loss degree is low, the current status can be maintained. If the loss degree is high, it can be processed and optimized in time, so that the logistics system in the storage area can reduce losses when handling express items.

[0053] 7. The image data obtained by the monitoring module is compared to realize the traceability of the target parts. After obtaining the flow process of the target parts, the location and corresponding area where the target parts are deformed and damaged can be determined. When it is necessary to optimize the warehouse management, the optimization target can be improved.

[0054] 8. By combining data recognition to obtain the corresponding optimization features, the logistics warehousing optimization knowledge graph gives the corresponding optimization plan. When optimizing warehousing management, targeted optimization can be achieved to improve the optimization effect. At the same time, by outputting the optimization plan as a reference, the optimization efficiency can be improved and the loss when classifying or transporting express items can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 The figure is a flow chart of a visual logistics tracing method based on a video monitoring network according to the present invention;

[0056] Figure 2 The figure is a schematic diagram of the structure of a visual logistics tracing system based on a video surveillance network according to the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] See also Figure 1 The present invention provides a visual logistics tracing method based on a video monitoring network, comprising:

[0059] Step 1: Collect express loss data in the storage area, and construct the loss degree Sts in the storage area based on the express loss data. If the loss degree Sts exceeds expectations, select risk nodes from several logistics nodes in the storage area based on the amount of express delivery passing through each logistics node;

[0060] The step 1 includes the following contents:

[0061] Step 101: before the express items are sent to the storage area of ​​the logistics center and need to be transported and distributed, the storage area is used as a monitoring area, and an electronic map covering the monitoring area is constructed; a sub-period including several monitoring periods is set, and the loss amount Su of the express items in the monitoring area is counted in each sub-period, and the loss amount data in each sub-area is aggregated to construct a data set of express item losses;

[0062] Step 102: construct the loss degree Sts from the express loss data set, wherein the loss amount Su is dimensionless and then calculated as follows:

[0063]

[0064] Where i = 1, 2, ..., p, p is the number of sub-periods, and the weight coefficient is: 0 ≤ F 1 ≤1,0≤F 2 ≤1 and F 2 +F 1 =1; Su i is the loss amount of express delivery in the ith sub-period, Su avg is the mean of the loss amount; is the acceptable value of loss amount, and the weight coefficient is obtained by referring to the hierarchical analysis method;

[0065] Based on historical data and management expectations for express parcels in the storage area, a loss threshold is pre-set; if the loss degree Sts exceeds the preset loss threshold, it means that the loss degree of the current express parcel is relatively serious and needs to be processed in time. At this time, a screening instruction is issued to the outside;

[0066] When in use, when express parcels are transported and distributed in the storage area, the loss degree of express parcels in each sub-period is counted, and the loss degree is constructed based on this. The loss degree is used to make a comprehensive judgment on the loss degree of express parcels in the storage area. If the loss degree is high, it means that the express parcels in the storage area need to be optimized during circulation;

[0067] Step 103: After receiving the screening instruction, the logistics nodes such as warehouses, sorting centers, loading and unloading areas, distribution centers, and handover points are marked in the monitoring area according to the logistics distribution process; the logistics nodes are marked on the electronic map, and the number of express items Tp passing through each logistics node is counted in the sub-period;

[0068] The risk level Fts is constructed based on the express delivery volume, and the risk level of each logistics node is evaluated based on the risk level Fts. After the express delivery volume Tp is dimensionless, the following method is used:

[0069]

[0070] Among them, F 1 、F 2 is the weight coefficient, 0≤F 1 ≤1,0≤F 2 ≤1, and F 1 +F 2 =1, i = 1, 2, ... p, p is the number of sub-periods, Tp i is the number of express parcels at the logistics node in the i-th sub-period, Tp avg is the mean of the express delivery volume; the weight coefficient is consistent with the previous value;

[0071] If the risk level Fts exceeds the preset risk threshold, the corresponding logistics node is used as the risk node, and after the risk node is marked on the electronic map, a supplementary instruction is issued to the outside;

[0072] When using, combine the contents in steps 101 to 103:

[0073] When optimizing or adjusting the logistics process within the storage area, the corresponding logistics nodes are determined and the risk level Fts at each logistics node is calculated. The risk nodes are screened out from several logistics nodes based on the risk level Fts. When optimizing and adjusting the transportation and distribution of express items, more obvious optimization goals can be achieved, and the processing efficiency and targeted processing can be improved during actual processing.

[0074] Step 2: Add a monitoring module at the risk node and detect the communication quality data in the vicinity of the logistics node. Use the acquired communication quality data to construct the communication coefficient Xts. If the communication coefficient Xts is lower than the quality threshold, switch the data transmission mode of the monitoring module.

[0075] The step 2 includes the following contents:

[0076] Step 201: After receiving the supplementary instruction, a monitoring module, such as a camera, is added at the risk node, and the installation position of the monitoring module is constrained in the following manner:

[0077]

[0078] Where n is the number of monitoring modules, V(i, j) is the distance from monitoring module i to monitoring module j, S avg is the mean distance, Fts is the risk of the logistics node, Tp is the number of express parcels passing through the logistics node, and the above parameters are dimensionless;

[0079] After completing the installation location and quantity constraints, the monitoring modules are installed and deployed. When adding monitoring modules at risk nodes, by increasing the number of monitoring modules, when collecting images of express items, the amount of image data collected can be increased, and the tracking accuracy can be improved when tracking express items;

[0080] Step 202: After installing the monitoring module at the risk node, the communication quality in the vicinity of the logistics node is monitored, and the acquired communication quality data is aggregated to construct a communication status data set;

[0081] Including: data transmission indicators, such as bit rate (bit / s): the number of binary bits of data transmitted per unit time in the communication system; baud rate (baud): the number of changes in the waveform of the transmitted signal per unit time in the communication system. Communication efficiency: the ratio of the data frame transmission time to the total time used to send the message; reliability indicators, such as bit error rate: an indicator to measure the reliability of a digital communication system, which is the probability that a binary code is transmitted incorrectly during the transmission process;

[0082] Step 203: Generate a communication coefficient Xts from the communication status data set, and use the communication coefficient to determine whether the signal quality at the logistics node is qualified. After linear normalization of the bit error rate Ym and the communication efficiency Tm, the corresponding data values ​​are mapped to the interval [0, 1], and then in the following manner:

[0083]

[0084] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1; the weight coefficient can be obtained by referring to the hierarchical analysis method;

[0085] If the communication coefficient Xts is lower than the preset quality threshold, it means that the current communication quality is poor. When the signal strength changes, different data transmission modes are switched, for example, data transmission between wired network and wireless network modes, so as to improve the current communication quality.

[0086] When used, combine the contents of steps 201 to 203;

[0087] After completing the installation and constraints of the monitoring module, the communication status in the area where the monitoring module is located is detected, and the corresponding communication coefficient Xts is constructed based on the acquired detection data. The communication capacity at the logistics node is evaluated and judged based on the communication coefficient Xts. If the communication capacity is good, the current data transmission mode can be maintained. If the current communication quality is poor, in this case, the current data transmission mode, that is, the communication mode, can be switched to ensure and adjust the current communication quality, so that after acquiring image data or other data, the data can be sent out in time.

[0088] Step 3: The installed recognition module monitors the flow of express items in the storage area, and uses the acquired monitoring data and related data to train and build a digital twin model of the express item flow, and uses real-time data to update the model;

[0089] The step three includes the following contents:

[0090] Step 301: When express items are transferred between various logistics nodes, identification modules are installed on the logistics nodes and corresponding transportation vehicles. The identification modules include hardware devices such as RFID, locators, and barcode scanners, which are used to record and track logistics information and obtain location information of each express item.

[0091] Step 302: The identification module collects the location data and corresponding time data of the express item, collects the equipment and related data when the express item is delivered and transferred, and aggregates the above collected data to construct a collection of express item transportation flow data;

[0092] Extract some data from the express delivery transportation flow data set as sample data, and build an initial model using a machine learning algorithm; select sample data to train the initial model, obtain the trained express delivery flow digital twin model, and use real-time data to update the model;

[0093] When using, combine the contents in steps 301 and 302:

[0094] After completing the installation of the monitoring module, when the express items are circulated, the monitoring module collects the corresponding data and trains to obtain the digital twin model of the express item circulation. When the express items are managed in the storage area, after obtaining the location and image data of the express items, by constructing the digital twin model of the express item circulation, the express item circulation process can be displayed, and the circulation and distribution of express items in the storage area can be predicted.

[0095] Step 4: Obtain image data of express items during circulation, and select target items based on the degree of deformation of express items at each logistics node through image recognition and comparison. Analyze the target item data at the logistics node to obtain the corresponding loss density Mj. If the loss density Mj is higher than the density threshold, issue a damaged item reminder.

[0096] The step 4 includes the following contents:

[0097] Step 401: The monitoring module acquires image data of the express item during its circulation at the logistics node, and constructs a corresponding image recognition model by a machine learning algorithm. The image recognition model performs object detection and recognition on the acquired image data, identifies the relevant data of the express item and the transportation tool, and determines the corresponding location information after the express item is identified.

[0098] Step 402: after the express item is identified, the timestamp and location information of the express item are used as matching items, and the data obtained by the identification module are matched according to the matching items; after the image data of each express item is obtained, the current image data of the express item is compared with the initial image data to determine the deformation degree of the express item at each logistics node, and the deformation degree is used as the deformation ratio. If the deformation ratio exceeds the expectation, the corresponding express item is determined as the target item;

[0099] When in use, images of the express items are captured while they are being transported, and through image recognition and comparison, each express item is screened according to its deformation. Items with a higher degree of deformation are selected as target items. By screening out the target items, they can be processed in a timely manner, reducing losses caused by product deformation or damage.

[0100] Step 403: mark the target item in the digital twin model of express item flow. After obtaining the target item data at each logistics node, such as the current location information, quantity and time information of the target item, analyze and obtain the corresponding loss density Mj under dimensionless conditions, as follows:

[0101]

[0102] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, the weight coefficient is determined by referring to the hierarchical analysis method; i=1, 2, …k, k is the number of target parts; D ijis the shortest distance from the i-th target part to the j-th target part, Df avg is the distance mean; W ij is the difference in flow speed between the i-th target part and the j-th target part, Wf avg is the mean value of flow velocity difference;

[0103] Based on historical data and the expectations for the flow management between express parcels and various logistics nodes, a density threshold is pre-set; if the loss density Mj is higher than the density threshold, it means that under the current circumstances, many express parcels are damaged in the monitoring area and thus suffer losses. When the express parcels are transported, circulated and distributed, some logistics links may have problems, which require timely control and issuance of damaged part reminders to the outside world;

[0104] When used, combine the contents in steps 401 to 403:

[0105] After obtaining the location and quantity of the target items, the corresponding loss density Mj is constructed. The loss degree of the express items in the current storage area is judged based on the loss density Mj. If the loss degree is low and still within an acceptable range, the current status can be maintained. If the loss degree is high, it can be processed and optimized in time, so that the logistics system in the storage area can reduce losses when handling express items.

[0106] Step 5: Construct the logistics information chain of the target item based on the acquired image data, determine the number of damaged express items in the sub-area based on the logistics information chain, and construct the damage coefficient Qts of the sub-area. If the damage coefficient Qts exceeds expectations, the logistics warehousing optimization knowledge graph will provide a corresponding optimization plan;

[0107] The step five includes the following contents:

[0108] Step 501: After receiving the damaged part reminder, the target part information is obtained, including the location and image information; on the basis of obtaining historical data, independent frames containing the target part information are extracted from the monitoring video, and several independent frames are connected according to the acquisition time to form a logistics information chain of the target part;

[0109] Step 502: Divide the monitoring area into several sub-areas. After the damaged position of the target item is determined by the logistics information chain, the number of damaged items in the sub-area in each sub-period is determined, and then the damage coefficient Qts of the sub-area is constructed, wherein the damage quantity Xq is linearly normalized and the corresponding data value is mapped to the interval [0, 1], and then according to the following formula:

[0110]

[0111] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, the weight coefficient is consistent with the previous value; i=1, 2, …, p, p is the number of sub-periods; Xq i is the number of damaged express items in the i-th sub-period, Xq avg is the mean number of damaged items, is the qualified standard value for the number of damaged items.

[0112] If the damage coefficient Qts exceeds expectations, it means that the corresponding sub-region may have certain abnormalities. The corresponding sub-region is regarded as an abnormal region. If the abnormal region accounts for a large proportion, exceeding the expected proportion, an optimization instruction is issued to the outside.

[0113] When in use, after the target part is determined, the image data obtained by the monitoring module is compared to realize the traceability of the target part. After obtaining the flow process of the target part, the position and corresponding area where the target part is deformed and damaged can be determined. When it is necessary to optimize the warehouse management, the optimization target can be improved.

[0114] Step 503: After receiving the optimization instruction, the monitoring data and express delivery flow data in the monitoring area are obtained, and the above data are identified and the corresponding optimization features are obtained in combination with the pre-set qualification standards. Logistics and warehousing optimization is used as the target word. After deep retrieval and entity relationship building, a logistics and warehousing optimization knowledge graph is pre-built;

[0115] Using the trained matching model, according to the correspondence between the optimization features and the optimization scheme, the logistics warehousing optimization knowledge graph gives the corresponding optimization scheme, for example, adjusting the transportation route of express parcels in the warehousing area, optimizing the warehousing layout, improving equipment performance, etc., and using the optimization scheme to optimize the express parcel flow process in the monitoring area;

[0116] When using, combine the contents in steps 501 to 503:

[0117] When it is determined that optimization is needed, the corresponding optimization features are obtained by combining data identification, and the corresponding optimization plan is given by the logistics warehousing optimization knowledge graph. When optimizing warehousing management, targeted optimization is achieved to improve the optimization effect. At the same time, by outputting the optimization plan as a reference, the efficiency of optimization is improved, the subsequent loss of express parcels is reduced, and the loss when classifying or transporting express parcels is reduced.

[0118] The construction method of logistics warehousing optimization knowledge graph can refer to the following:

[0119] Clarify goals and requirements: Determine the specific application scenarios of knowledge graphs in logistics and warehousing optimization, such as improving transportation efficiency, reducing costs, optimizing inventory management, etc. Clarify the entities, relationships, and attributes that need to be described, such as goods, warehouses, orders, transportation routes, etc.

[0120] Data collection and preprocessing: Collect warehousing-related data from various structured and unstructured data sources (such as ERP systems, WMS systems, order systems, etc.). Perform preprocessing operations such as cleaning, deduplication, and format conversion on the data to ensure data quality and consistency.

[0121] Knowledge graph construction: Entity recognition: Use named entity recognition technology to extract warehousing-related entities from the data, such as goods, orders, warehouses, etc.

[0122] Relation extraction: Use relationship extraction technology to extract various semantic relationships between entities from the data, such as goods stored in warehouses, orders associated with goods, etc.

[0123] Attribute extraction: Extract attributes from entities, such as the weight, size, production date, etc. of the product.

[0124] Knowledge fusion: Due to the diversity of data sources, there may be problems such as knowledge duplication and unclear relationships. Knowledge fusion is performed through technologies such as entity alignment and attribute value filling to eliminate ambiguity and form a unified knowledge system.

[0125] Knowledge graph storage and query: Choose a suitable graph database (such as Neo4j, JanusGraph, etc.) to store the knowledge graph. Design efficient query algorithms and interfaces to support query and analysis of the knowledge graph.

[0126] Knowledge graph application and optimization: Apply the constructed knowledge graph to actual scenarios of logistics warehousing optimization, such as inventory forecasting, route planning, order processing, etc. Provide feedback and optimization based on the application results, and continuously improve the construction method and application strategy of the knowledge graph.

[0127] Iterative updates and maintenance: As logistics and warehousing business develops and changes, the knowledge graph is continuously updated to ensure that it keeps pace with actual application scenarios. The knowledge graph is regularly maintained and optimized to improve query efficiency and accuracy.

[0128] The Analytic Hierarchy Process (AHP) is a decision-making method that decomposes decision-related elements into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. It is particularly suitable for dealing with target systems with hierarchical and staggered evaluation indicators, and when the target value is difficult to describe quantitatively, the AHP is an effective decision-making tool.

[0129] The core of the hierarchical analysis method is to decompose the decision problem into multiple levels to form a hierarchical structure, which usually includes the target level, the criterion level, the sub-criterion level and the solution level. By solving the eigenvector of the judgment matrix, the priority weight of each element of each level to an element of the previous level is obtained, and the final weight of each alternative solution to the total goal is recursively merged by the weighted sum method, so as to find the optimal solution.

[0130] See also Figure 2 The present invention provides a visual logistics tracing system based on a video monitoring network, comprising:

[0131] The risk analysis unit collects express loss data in the storage area and constructs the loss degree Sts in the storage area based on the express loss data. If the loss degree Sts exceeds expectations, the risk nodes are screened out from several logistics nodes in the storage area based on the amount of express delivery passing through each logistics node.

[0132] The communication quality detection unit adds a monitoring module at the risk node and detects the communication quality data in the vicinity of the logistics node. The communication coefficient Xts is constructed based on the acquired communication quality data. If the communication coefficient Xts is lower than the quality threshold, the data transmission mode of the monitoring module is switched.

[0133] The model building unit uses the installed recognition module to monitor the flow of express items in the storage area, and uses the acquired monitoring data and related data to train and build a digital twin model of the express item flow, and uses real-time data to update the model;

[0134] The loss monitoring unit obtains image data of express items during circulation, and selects target items based on the degree of deformation of express items at each logistics node through image recognition and comparison. The corresponding loss density Mj is obtained by analyzing the target item data at the logistics node. If the loss density Mj is higher than the density threshold, a damaged item reminder is issued.

[0135] The logistics optimization unit constructs the logistics information chain of the target item based on the acquired image data. The logistics information chain determines the number of damaged express items in the sub-area and constructs the damage coefficient Qts of the sub-area. If the damage coefficient Qts exceeds expectations, the logistics warehousing optimization knowledge graph gives a corresponding optimization plan.

[0136] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments 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 or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0137] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0138] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0139] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0140] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0141] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0142] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0143] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A visual logistics tracing method based on a video surveillance network, characterized by: include, Collect the loss data of express items in the storage area, and construct the loss degree in the storage area based on the loss data of express items If the loss If the number of express deliveries exceeds expectations, risk nodes are selected from several logistics nodes in the storage area according to the express delivery volume passing through each logistics node; a sub-cycle containing several monitoring cycles is set, among which the loss degree The construction method is as follows: After dimensionless processing, follow the following method: ； Weight coefficient: , and ; , is the number of sub-periods, For the The amount of express delivery loss in a sub-period, is the mean of the loss amount; The acceptable value of loss for the amount of loss; Add a monitoring module at the risk node and detect the communication quality data in the vicinity of the logistics node. Use the acquired communication quality data to construct the communication coefficient. , if the communication coefficient If the quality is lower than the threshold, the data transmission mode of the monitoring module is switched; after receiving the supplementary instruction, a monitoring module is added at the risk node, and the installation position of the monitoring module is constrained. The constraint method is as follows: ； in, is the number of monitoring modules, It is a monitoring module To the monitoring module The distance is the mean distance, is the risk of the logistics node, The number of express parcels passing through the logistics node; The communication quality in the vicinity of the logistics node is monitored, and the acquired communication quality data is aggregated to construct a communication status data set; the communication coefficient is generated from the communication status data set. , where the bit error rate and communication efficiency After linear normalization, the corresponding data values ​​are mapped to the interval Then follow the steps below: ； Weight coefficient: , ; If the communication coefficient When the quality is lower than the preset threshold, different data transmission modes are switched when the signal strength changes; The installed identification module monitors the flow of express items in the storage area, and uses the acquired monitoring data and related data to train and build a digital twin model of express item flow, and uses real-time data to update the model; Obtain image data of express parcels during their circulation, and through image recognition and comparison, select target parcels based on the degree of deformation of parcels at each logistics node, and obtain the corresponding loss density based on the data analysis of target parcels at the logistics nodes. , if the loss density If the density is higher than the threshold, a damaged item reminder is issued; the target item is marked in the digital twin model of the express delivery flow, and after obtaining the target item data at each logistics node, the corresponding loss density is analyzed under dimensionless conditions. , as follows: ； Weight coefficient: , ,and , , is the number of target pieces; It is Target items to The shortest distance between the target objects, is the distance mean; It is Target items to The difference in the flow speed of the target parts, is the mean value of flow velocity difference; The logistics information chain of the target item is constructed based on the acquired image data. The number of damaged items in the sub-area is determined based on the logistics information chain and the damage coefficient of the sub-area is constructed. , if the damage coefficient If the damage exceeds the expectation, the corresponding optimization plan will be given by the logistics warehousing optimization knowledge map; the damage coefficient The method of obtaining is as follows: Perform linear normalization and map the corresponding data values ​​to the interval Then follow the following formula: ； Weight coefficient: , ,and , , is the number of sub-periods; For the The number of damaged express items in a sub-period, is the mean number of damaged items, is the qualified standard value for the number of damaged items.

2. The visual logistics tracing method based on video surveillance network according to claim 1 is characterized in that: The storage area is used as the monitoring area, and the loss of express items in the monitoring area is counted in each sub-period. , after summarizing the loss data in each sub-area, a data set of express delivery loss is constructed; the loss degree is constructed from the express delivery loss data set If the loss When the preset loss threshold is exceeded, a screening instruction is issued to the outside.

3. The visual logistics tracing method based on video surveillance network according to claim 2 is characterized in that: After receiving the screening instruction, the logistics nodes are marked in the monitoring area, the logistics nodes are marked on the electronic map, and the number of express parcels passing through each logistics node is counted in the sub-period. ; Construct risk level based on express delivery volume , risk The construction method is as follows: After dimensionless processing, follow the following method: ； in, is the weight coefficient, , ,and , , is the number of sub-periods, For the The number of express parcels at the logistics node in the sub-period, is the mean of the express delivery volume; If the risk If the preset risk threshold is exceeded, the corresponding logistics node will be used as a risk node and additional instructions will be issued to the outside.

4. The visual logistics tracing method based on video surveillance network according to claim 1 is characterized by: When express items are transferred between various logistics nodes, identification modules are installed on the logistics nodes and corresponding transportation vehicles to record and track logistics information and obtain the location information of each express item; The identification module collects the location data and corresponding time data of the express items, collects the equipment and related data when the express items are distributed and transferred, and then summarizes them to build a collection of express item transportation flow data.

5. The visual logistics tracing method based on video surveillance network according to claim 4 is characterized in that: The monitoring module acquires image data of express items in circulation at the logistics node, and the image recognition model performs object detection and recognition on the acquired image data, identifies the relevant data of the express items and transportation tools, and determines the corresponding location information after identifying the express items; The current image data of the express item is compared with the initial image data to determine the degree of deformation of the express item at each logistics node. The degree of deformation is used as the deformation ratio. If the deformation ratio exceeds expectations, the corresponding express item will be taken as the target item.

6. The visual logistics tracing method based on video surveillance network according to claim 5 is characterized in that: After receiving the damaged part reminder, the target part information is obtained, and the independent frames containing the target part information are extracted from the monitoring video. Several independent frames are connected according to the acquisition time to form the logistics information chain of the target part; the monitoring area is divided into several sub-areas, and after the damaged position of the target part is determined by the logistics information chain, the number of damaged express parts in the sub-area in each sub-period is determined, and the damage quantity coefficient of the sub-area is constructed , if the damage coefficient If the ratio exceeds the expected value, the corresponding sub-region will be regarded as an abnormal region. If the ratio of abnormal regions exceeds the expected proportion, an optimization instruction will be issued to the outside. After receiving the optimization instruction, the corresponding optimization features are identified and acquired, and the logistics warehousing optimization is used as the target word to pre-build the logistics warehousing optimization knowledge graph; According to the correspondence between the optimization features and the optimization scheme, the corresponding optimization scheme is given by the logistics warehousing optimization knowledge graph, and the optimization scheme is used to optimize the express delivery flow process in the monitoring area.

7. A visual logistics tracing system based on a video surveillance network, applying the tracing method according to any one of claims 1 to 6, characterized in that: include: The risk analysis unit collects the loss data of express items in the storage area and constructs the loss degree in the storage area based on the loss data of express items. If the loss Exceeding expectations, based on the express delivery volume passing through each logistics node, risk nodes are screened out from several logistics nodes within the storage area; Communication quality detection unit: add a monitoring module at the risk node, detect the communication quality data in the vicinity of the logistics node, and construct the communication coefficient based on the acquired communication quality data. , if the communication coefficient If the quality is lower than the threshold, the data transmission mode of the monitoring module is switched; The model building unit uses the installed recognition module to monitor the flow of express items in the storage area, and uses the acquired monitoring data and related data to train and build a digital twin model of the express item flow, and uses real-time data to update the model; The loss monitoring unit obtains image data of express parcels during their circulation, and through image recognition and comparison, selects target parcels based on the degree of deformation of the parcels at each logistics node, and obtains the corresponding loss density based on the data analysis of the target parcels at the logistics node. , if the loss density If the density is higher than the threshold, a damaged parts reminder will be issued; The logistics optimization unit constructs the logistics information chain of the target item based on the acquired image data, determines the number of damaged express items in the sub-area based on the logistics information chain, and constructs the damage coefficient of the sub-area , if the damage coefficient Exceeding expectations, the corresponding optimization plan was given by the logistics and warehousing optimization knowledge graph.

Images