Deliver helmet wearing management method, device and equipment and storage medium

Abstract

The invention relates to the field of logistics, and discloses a courier helmet wearing management method, device and equipment and a storage medium, and the method is used for carrying out the detection management of the helmet wearing condition of a courier, and creating a key attention list for the courier who does not pass the sampling inspection. The method comprises the following steps: collecting real-time images of exits and entrances of express sites, and naming the real-time images; classifying the named images according to names of couriers, and storing the images in a to-be-detected database; obtaining a sampling inspection list, obtaining an image corresponding to the courier in the sampling inspection list from the to-be-detected database as a sampling inspection image, and carrying out sampling inspection on the sampling inspection image to obtain a sampling inspection result; screening couriers who do not pass the sampling inspection from the sampling inspection result, and generating notification information corresponding to the couriers; and collecting a helmet wearing appeal list, screening out couriers who do not pass appeal from an audit result of the helmet wearing appeal list, and listing the couriers who do not pass appeal into a pre-created key attention list.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and in particular to a method, device, equipment and storage medium for managing the wearing of helmets by couriers. Background Art

[0002] Currently, couriers need to wear helmets for safety reasons to reduce the risk of riding during the delivery process. However, some couriers are unable to wear helmets as required. At the same time, there is no complete system to monitor the couriers' current helmet wearing status, making it impossible to effectively supervise them.

[0003] Therefore, the existing technology still needs to be improved and developed. Summary of the invention

[0004] The present invention provides a method, device, equipment and storage medium for managing the wearing of helmets by couriers, which are used to detect and manage the wearing of helmets by couriers and create a key focus list for couriers who fail the random inspection.

[0005] The first aspect of the present invention provides a method for managing the wearing of helmets by couriers, the method comprising: collecting real-time images of the entrances and exits of express delivery sites, and naming the real-time images in the form of a name combined with a serial number after performing face recognition on the collected real-time images; classifying the named images according to the names of the couriers, and saving the classified images in a database to be detected; obtaining a random inspection list, and obtaining images corresponding to the couriers in the random inspection list from the database to be detected as random inspection images, and performing random inspections on the random inspection images to obtain random inspection results; screening out couriers who have failed the random inspection from the random inspection results, and generating notification information corresponding to the couriers; collecting helmet wearing complaint forms, and screening out couriers whose complaints have not been passed from the review results of the helmet wearing complaint forms, and placing the couriers whose complaints have not been passed in a pre-created key attention list.

[0006] Optionally, in a first implementation method of the first aspect of the present invention, the real-time images of the entrances and exits of express delivery stations are collected, and after performing face recognition on the collected real-time images, the real-time images are named by combining names with serial numbers, including: collecting real-time images of the entrances and exits of express delivery stations, and preprocessing the collected real-time images; performing face recognition on the preprocessed real-time images using a face recognition algorithm, and eliminating images in which no faces are recognized or the facial features are unclear, to obtain images to be named; and naming the images to be named by combining names with serial numbers according to the face recognition results.

[0007] Optionally, in a second implementation method of the first aspect of the present invention, classifying the named images according to the couriers' names and saving the classified images in a database to be detected includes: using regular expressions to classify the named images according to the couriers' names; creating a courier image folder corresponding to the courier's name in the database to be detected; and saving the classified images to the corresponding courier image folder based on the classification results.

[0008] Optionally, in a third implementation method of the first aspect of the present invention, obtaining a spot check list, obtaining images corresponding to the couriers in the spot check list from the database to be detected as spot check images, and spot checking the spot check images to obtain spot check results, includes: obtaining a spot check list, using a keyword matching algorithm to obtain images corresponding to the couriers in the spot check list from the database to be detected as spot check images; for each of the spot check images, inputting the spot check image into an improved detection model, and obtaining a predicted helmet position output by the improved detection model; determining whether the courier is wearing the helmet correctly based on the predicted helmet position, and if so, outputting the spot check result that the courier is wearing the helmet correctly, otherwise, outputting the spot check result that the courier is not wearing the helmet correctly.

[0009] Optionally, in a fourth implementation manner of the first aspect of the present invention, for each of the sampled images, the sampled image is input into the improved detection model, and the predicted helmet position output by the improved detection model is obtained, and the method also includes: constructing the basic structure of the YOLOv model, the basic structure of the YOLOv model including a backbone network, a feature detection layer, a rough position prediction layer and a precise position prediction layer; adding a CBAM attention module to the backbone network and the feature detection layer respectively to obtain an updated YOLOv model, modifying the anchor frame calculation algorithm of the updated YOLOv model to a differential evolution algorithm, and introducing the mutation and selection mechanism in the genetic algorithm to optimize the size and shape of the anchor frame, thereby obtaining an improved YOLOv model; collecting and annotating image data containing couriers wearing helmets as a training set, and using the training set to train and optimize the improved YOLOv model to obtain an improved detection model.

[0010] Optionally, in a fifth implementation method of the first aspect of the present invention, the method of screening out couriers who fail the random inspection from the random inspection results and generating notification information corresponding to the couriers includes: statistics on the random inspection results, calculating the random inspection pass rate, and generating a random inspection report based on the statistical results and the random inspection pass rate; identifying the couriers who fail the random inspection from the random inspection report, and obtaining communication information of the couriers who fail the random inspection; calling a preset notification template, filling the courier's name into the preset notification template, generating notification information, and sending the notification information to the courier based on the obtained communication information.

[0011] Optionally, in a sixth implementation method of the first aspect of the present invention, the collecting of helmet wearing complaint forms, and screening out the couriers whose appeals have not been approved from the review results of the helmet wearing complaint forms, and placing the couriers whose appeals have not been approved on a pre-created key attention list comprises: collecting helmet wearing complaint forms, and screening out the helmet wearing complaint forms that have completed the review according to the status of the helmet wearing complaint forms; dividing the couriers in the helmet wearing complaint forms that have completed the review into two categories: passed and failed according to the review results of the helmet wearing complaint forms; creating a key attention list, and placing the couriers whose appeals have not been approved on the pre-created key attention list.

[0012] The second aspect of the present invention provides a courier helmet wearing management device, including: a naming module, used to collect real-time images of the entrances and exits of express delivery sites, and name the real-time images in the form of a name combined with a serial number after performing face recognition on the collected real-time images; a saving module, used to classify the named images according to the names of the couriers, and save the classified images in a database to be detected; a detection module, used to obtain a random inspection list, and obtain images corresponding to the couriers in the random inspection list from the database to be detected as random inspection images, and perform random inspections on the random inspection images to obtain random inspection results; a statistical module, used to filter out couriers who have failed the random inspection from the random inspection results, and generate notification information corresponding to the couriers; a management module, used to collect helmet wearing complaint forms, and filter out couriers whose complaints have not been passed from the review results of the helmet wearing complaint forms, and include the couriers whose complaints have not been passed in a pre-created key attention list.

[0013] Optionally, in a first implementation method of the second aspect of the present invention, the naming module includes: a preprocessing unit, which collects real-time images of the entrances and exits of the express station and preprocesses the collected real-time images; a first recognition unit, which uses a face recognition algorithm to perform face recognition on the preprocessed real-time images, and eliminates images in which no face is recognized or the facial features are unclear, to obtain images to be named; and a naming unit, which names the images to be named with a name combined with a serial number according to the face recognition results.

[0014] Optionally, in a second implementation of the second aspect of the present invention, in this embodiment, the saving module includes: a first classification unit, used to classify the named images according to the courier's name using a regular expression; a first creation unit, used to create a courier image folder corresponding to the courier's name in the database to be detected; and a saving unit, used to save the classified image to the corresponding courier image folder according to the classification result.

[0015] Optionally, in a third implementation of the second aspect of the present invention, the detection module includes: an acquisition unit, used to obtain a random inspection list, and use a keyword matching algorithm to obtain images corresponding to the couriers in the random inspection list from the database to be detected as random inspection images; a detection unit, used to input the random inspection image into an improved detection model for each of the random inspection images, and obtain the predicted helmet position output by the improved detection model; a judgment unit, used to judge whether the courier is wearing the helmet correctly according to the predicted helmet position, and when it is judged that the courier is wearing the helmet correctly, outputting the random inspection result that the courier is wearing the helmet correctly, otherwise, outputting the random inspection result that the courier is not wearing the helmet correctly.

[0016] Optionally, in a fourth implementation method of the second aspect of the present invention, the statistical module includes: a statistical unit, used to count the inspection results, calculate the inspection pass rate, and generate an inspection report based on the statistical results and the inspection pass rate; a second identification unit, used to identify couriers who failed the inspection from the inspection report, and obtain communication information of the couriers who failed the inspection; a notification unit, used to call a preset notification template, fill in the courier's name into the preset notification template, generate notification information, and send the notification information to the courier based on the obtained communication information.

[0017] Optionally, in a fifth implementation of the second aspect of the present invention, the management module includes: a screening unit, used to collect helmet wearing complaint forms, and screen out the helmet wearing complaint forms that have been reviewed according to the status of the helmet wearing complaint forms; a second classification unit, used to divide the couriers in the helmet wearing complaint forms that have been reviewed into two categories: passed and failed according to the review results of the helmet wearing complaint forms; and a second creation unit, used to create a key attention list, and include the couriers whose complaints have not been passed in the key attention list.

[0018] The third aspect of the present invention provides a courier helmet wearing management device, comprising: a memory and at least one processor, the memory storing computer-readable instructions, the memory and the at least one processor being interconnected through lines; the at least one processor calls the computer-readable instructions in the memory so that the courier helmet wearing management device executes the various steps of the courier helmet wearing management method as described above.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the various steps of the courier helmet wearing management method as described above.

[0020] In the technical solution provided by the present invention, through real-time image collection and face recognition technology, the identity of the courier and the image naming are automatically carried out, and the named images are classified according to the couriers' names and saved in the database to be detected. When the random inspection is carried out, the images corresponding to the couriers in the random inspection list are obtained from the database to be detected according to the random inspection list for random inspection, and after the random inspection is completed, the random inspection results are counted, so as to realize real-time monitoring and analysis of the couriers' helmet wearing situation, generate notification information for the couriers who fail the random inspection, collect helmet wearing complaint forms, and screen out the couriers whose complaints are not passed from the review results of the helmet wearing complaint forms, and put the couriers whose complaints are not passed into the pre-created key attention list, and strengthen the management of the couriers on the key attention list, which is conducive to improving the safety protection level of the delivery personnel and reducing the injuries caused by traffic accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A first flow chart of a method for managing courier helmet wearing provided by an embodiment of the present invention;

[0022] Figure 2 A second flow chart of the courier helmet wearing management method provided in an embodiment of the present invention;

[0023] Figure 3 A third flow chart of the courier helmet wearing management method provided in an embodiment of the present invention;

[0024] Figure 4 A fourth flow chart of the courier helmet wearing management method provided by an embodiment of the present invention;

[0025] Figure 5 A fifth flow chart of the courier helmet wearing management method provided in an embodiment of the present invention;

[0026] Figure 6 A sixth flow chart of the courier helmet wearing management method provided by an embodiment of the present invention;

[0027] Figure 7 A schematic diagram of the structure of a courier helmet wearing management device provided by an embodiment of the present invention;

[0028] Figure 8 A schematic diagram of the structure of a courier helmet wearing management device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The embodiment of the present invention provides a method, device, equipment and storage medium for managing the wearing of helmets by couriers. The method is used to detect and manage the wearing of helmets by couriers, and to create a key focus list for couriers who fail the random inspection. The method includes: collecting real-time images of the entrances and exits of express delivery stations, and naming the real-time images in the form of names combined with serial numbers after face recognition of the collected real-time images; classifying the named images according to the names of the couriers, and saving the classified images in a database to be detected; obtaining a random inspection list, and obtaining images corresponding to the couriers in the random inspection list from the database to be detected as random inspection images, and performing random inspections on the random inspection images to obtain random inspection results; screening out couriers who fail the random inspection from the random inspection results, and generating notification information corresponding to the couriers; collecting helmet wearing complaint forms, and screening out couriers who fail the complaint from the review results of the helmet wearing complaint forms, and listing the couriers who fail the complaint in a pre-created key focus list.

[0030] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 A first embodiment of a method for managing courier helmet wearing in an embodiment of the present invention includes:

[0032] S101, collecting real-time images of the entrance and exit of the express delivery station, and naming the real-time images by combining the name with the serial number after performing face recognition on the collected real-time images.

[0033] In this embodiment, high-definition cameras are installed at the entrances and exits of the express delivery stations, which can capture images of couriers entering and leaving in real time.

[0034] In this embodiment, after collecting real-time images of the entrance and exit of the express delivery station, face recognition technology is used to perform face recognition on the captured real-time images to identify the identity of the courier. The identified courier's name is combined with the time and location of image capture to generate a unique serial number. Then, the image is named in the form of "name + serial number" to facilitate subsequent management and query.

[0035] S102: classify the named images according to the names of the couriers, and save the classified images in a database to be detected.

[0036] In this embodiment, the named images are classified according to the couriers' names, ensuring that the image of each courier is stored separately.

[0037] Establish a database to be tested and save all classified images in it for subsequent random inspection.

[0038] S103, obtaining a random inspection list, and obtaining images corresponding to the couriers in the random inspection list from the database to be inspected as random inspection images, and performing random inspection on the random inspection images to obtain random inspection results.

[0039] In this embodiment, the inspection list can set different inspection numbers according to the risk level of the express delivery site, and then select specific inspection objects by random sampling based on the determined inspection number.

[0040] In this embodiment, the YOLOv model can be used to automatically detect the sampled images to determine whether the courier is wearing a helmet. The YOLOv series model can quickly and accurately identify multiple target objects in an image.

[0041] S104. Filter out the couriers who failed the random inspection from the random inspection results, and generate notification information corresponding to the couriers.

[0042] In this embodiment, the sampling inspection results are statistically analyzed to calculate the sampling inspection pass rate, violation rate and other indicators. Then, for the couriers who fail the sampling inspection, notification information is automatically generated, including the violation time, location, violation type and other information. Finally, the notification information is sent to the relevant couriers and managers so that timely measures can be taken to rectify the situation.

[0043] S105. Collect helmet wearing complaint forms, and screen out couriers whose complaints have not been approved from the review results of the helmet wearing complaint forms, and put the couriers whose complaints have not been approved into a pre-created key focus list.

[0044] In this embodiment, a helmet wearing complaint channel is set up in advance to allow couriers to appeal against the inspection results.

[0045] Collect helmet wearing complaint forms. After the helmet wearing complaint forms are reviewed, screen out the couriers whose complaints have not been approved based on the review results of the helmet wearing complaint forms.

[0046] In this embodiment, a focus list is created, and the information of couriers whose appeals have not been approved is entered into the list. The couriers on the focus list are continuously tracked and supervised to ensure that they comply with company regulations and safety standards. For example, the frequency of random inspections of couriers on the focus list is increased.

[0047] In this embodiment, the key focus list is updated and adjusted regularly to ensure the accuracy and effectiveness of management.

[0048] It is understandable that the execution subject of the present invention may be a courier helmet wearing management device, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking the server as the execution subject as an example.

[0049] The present embodiment provides a method for managing the wearing of helmets by couriers, which automatically identifies the couriers and names their images through real-time image collection and face recognition technology, and classifies the named images according to the couriers' names and saves them in a database to be detected. When spot checks are carried out, images corresponding to the couriers in the spot check list are obtained from the database to be detected for spot checks based on the spot check list, and after the spot checks are completed, the spot check results are counted, thereby achieving real-time monitoring and analysis of the couriers' helmet wearing conditions. For couriers who fail the spot check, notification information is generated, and helmet wearing complaint forms are collected. Couriers whose complaints have not been passed are screened out from the review results of the helmet wearing complaint forms, and the couriers whose complaints have not been passed are included in a pre-created key focus list. Management of the couriers on the key focus list can be strengthened, which is beneficial to improving the safety protection level of delivery personnel and reducing injuries caused by traffic accidents.

[0050] See also Figure 2 , a second embodiment of the courier helmet wearing management method in the embodiment of the present invention includes:

[0051] S201, collecting real-time images of the entrances and exits of express delivery stations, and preprocessing the collected real-time images.

[0052] In this embodiment, the camera is connected to the image acquisition system, and the captured images are transmitted to the system in real time for subsequent processing.

[0053] In this embodiment, pre-processing operations such as denoising and contrast enhancement are performed on the received real-time image to improve image quality and facilitate subsequent face recognition.

[0054] Preprocessing also includes operations such as cropping and scaling the image to make it meet the input requirements of the face recognition algorithm.

[0055] S202, performing face recognition on the preprocessed real-time image using a face recognition algorithm, and eliminating images in which faces are not recognized or face features are unclear, to obtain images to be named.

[0056] In this embodiment, a face recognition algorithm is called to perform face detection and recognition on the preprocessed image.

[0057] The facial recognition algorithm analyses the facial features in the image and compares it with a pre-established facial database to identify the person in the image, i.e. the courier’s name.

[0058] In this embodiment, images in which no human face is recognized or the facial features are unclear are marked and discarded to ensure the accuracy of subsequent processing.

[0059] S203, naming the image to be named by combining the name with the serial number according to the face recognition result.

[0060] In this embodiment, for images that successfully identify a face, the system will extract the courier's name based on the face recognition results. In order to ensure the uniqueness and traceability of the image, a unique serial number is generated for each image. Finally, the image will be named in the form of "name + serial number" and saved in the corresponding database for subsequent management and retrieval.

[0061] For example, the images are named Zhangsan_20230401_001.jpg, Lisi_20230401_002.png, and Wangwu_Send_20230401_003.jpeg.

[0062] In this embodiment, real-time images of the entrances and exits of express delivery stations are collected, and after pre-processing the collected real-time images, automatic naming and classification of the images are achieved through face recognition technology, which provides a solid foundation for subsequent spot checks, statistics and analysis, and effectively improves the level of intelligent safety management of express delivery stations.

[0063] See also Figure 3 A third embodiment of a courier helmet wearing management method in an embodiment of the present invention includes:

[0064] S301. Classify the named images according to the names of the couriers using regular expressions.

[0065] In this embodiment, the named image file name includes the courier's name and the serial number. In order to accurately classify the images according to the courier's name, a regular expression can be used to parse the file name and extract the courier's name.

[0066] S302: Create a courier image folder corresponding to the courier's name in the database to be detected.

[0067] In this embodiment, in the database to be detected, an independent image folder is established for each courier to store all images of the courier. The folder name can correspond to the courier's name to facilitate subsequent management and retrieval.

[0068] S303. According to the classification result, the classified image is saved in a corresponding courier image folder.

[0069] In this embodiment, after the image classification is completed, each image needs to be saved in a corresponding courier image folder, and each image is saved in a corresponding folder according to the corresponding courier name by traversing the classified image list.

[0070] In this embodiment, the named images are automatically classified according to the identity of the couriers and are stored in an orderly manner in the database to be detected, thereby improving the efficiency and accuracy of image management and providing convenience for subsequent helmet wearing detection work.

[0071] See also Figure 4 A fourth embodiment of a courier helmet wearing management method in an embodiment of the present invention includes:

[0072] S401, obtaining a random inspection list, and using a keyword matching algorithm to obtain images corresponding to the couriers in the random inspection list from a database to be inspected as random inspection images.

[0073] In this embodiment, the inspection list can be provided by the relevant department or management personnel and then uploaded to the system, or the system can automatically generate the name or number of the courier to be inspected according to the preset rules. The inspection list can be provided in the form of a text file, database query results, etc.

[0074] S402: For each sampled image, input the sampled image into the improved detection model, and obtain the predicted position of the helmet output by the improved detection model.

[0075] In this embodiment, for each sampled image, the sampled image is input into the improved detection model, and the detection result output by the improved detection model is obtained, which also includes: constructing the basic structure of the YOLOv model, the basic structure of the YOLOv model includes a backbone network, a feature detection layer, a rough position prediction layer and a precise position prediction layer; adding a CBAM attention module to the backbone network and the feature detection layer respectively to obtain an updated YOLOv model, modifying the anchor frame calculation algorithm of the updated YOLOv model to a differential evolution algorithm, and introducing the mutation and selection mechanism in the genetic algorithm to optimize the size and shape of the anchor frame, thereby obtaining an improved YOLOv model; collecting and annotating image data containing couriers wearing helmets as a training set, and using the training set to train and optimize the improved YOLOv model to obtain an improved detection model.

[0076] In this embodiment, the backbone network is responsible for extracting feature information from the image, which is usually implemented using a deep convolutional neural network (CNN). The feature detection layer further processes the features extracted by the backbone network to detect the target in the image. The rough position prediction layer predicts the approximate position of the target in the image based on the results of the feature detection layer. The precise position prediction layer fine-tunes the rough position to obtain the precise position of the target in the image.

[0077] In this embodiment, the CBAM attention mechanism consists of two main parts: channel attention and spatial attention. It can adaptively adjust the importance of different channels and spatial positions, thereby improving the model's ability to express image features, helping the model to better capture key information in the image, and improving the accuracy and robustness of target detection.

[0078] In this embodiment, in the YOLOv model, a CBAM module can be added at the end of each residual unit of the backbone network, after the SPP (Spatial Pyramid Pooling) layer of the middle layer, and in the detection head, so that the model can better capture the dependencies between features and improve the feature representation capability and detection accuracy.

[0079] In this embodiment, the differential evolution algorithm is an optimization algorithm based on swarm intelligence theory, which realizes intelligent optimization search through cooperation and competition among individuals.

[0080] Anchor box calculation includes the following steps:

[0081] Initialize the population: Use the differential evolution algorithm to initialize the population of anchor boxes, that is, generate a set of initial anchor box sizes.

[0082] Fitness evaluation: The fitness of each anchor box is evaluated based on the degree of match between the anchor box and the true target box (such as IOU value).

[0083] Differential mutation: The anchor box is mutated through the differential mutation operator to generate a new anchor box.

[0084] Crossover operation: Perform a crossover operation on the mutated anchor boxes to generate a richer set of anchor box candidates.

[0085] Selection operation: Select the best anchor box as part of the next generation population based on fitness.

[0086] Iterative optimization: Repeat the above steps until the preset number of iterations is reached or the fitness is no longer significantly improved.

[0087] In this embodiment, on the basis of the differential evolution algorithm, the mutation and selection mechanism in the genetic algorithm can be introduced to further optimize the size and shape of the anchor frame.

[0088] In this embodiment, a keyword matching algorithm (such as exact matching, fuzzy matching, etc.) is used to search for images matching the courier name or number in the spot check list in the database file name or metadata. The matched image is extracted as a spot check image and prepared for the next spot check. After the spot check image is input into the detection model, the model outputs the predicted position of the helmet in the image. The predicted position is usually expressed in the form of a bounding box, which contains the coordinates and size information of the helmet in the image.

[0089] S403. Determine whether the courier wears the helmet correctly based on the predicted position of the helmet. If so, output the sampling result that the courier wears the helmet correctly. Otherwise, output the sampling result that the courier does not wear the helmet correctly.

[0090] In this embodiment, based on the predicted position of the helmet and the courier information in the image (such as head position, posture, etc.), it is determined whether the courier is wearing a helmet.

[0091] If the predicted position of the helmet matches the position of the courier's head, and the helmet is worn correctly (such as the helmet completely covers the head, the fastening strap is fastened, etc.), the helmet is judged to be worn correctly.

[0092] On the contrary, if the predicted position of the helmet is inaccurate, missing, or does not match the courier's head, it is determined that the helmet is not worn correctly.

[0093] In this embodiment, the random inspection result information is output according to the determination result. If the courier wears the helmet correctly, the random inspection result of "the courier wears the helmet correctly" is output.

[0094] If the courier does not wear a helmet correctly, the random inspection result of "the courier does not wear a helmet correctly" is output, and relevant image evidence or recommended measures may be attached.

[0095] In this embodiment, a keyword matching algorithm is used to accurately obtain images of couriers on the inspection list, and an improved detection model is used to efficiently and accurately determine whether the couriers are wearing helmets, thereby realizing the automation and intelligence of the inspection process and improving the inspection efficiency. Moreover, the improved detection model combined with the differential evolution algorithm and the attention mechanism can improve the detection accuracy.

[0096] See also Figure 5 A fifth embodiment of a method for managing courier helmet wearing in an embodiment of the present invention includes:

[0097] S501. Perform statistics on the sampling inspection results, calculate the sampling inspection pass rate, and generate a sampling inspection report based on the statistical results and the sampling inspection pass rate.

[0098] In this embodiment, the detection results of all random inspection images are collected, including the courier's name, whether the random inspection is passed, etc. The random inspection results are counted, and the random inspection pass rate is calculated, that is, the proportion of the number of couriers who passed the random inspection to the total number of random inspections. The random inspection pass rate can be used as an important indicator to evaluate the helmet wearing condition of couriers.

[0099] In this embodiment, a random inspection report is generated based on the statistical results and the random inspection pass rate. The random inspection report includes key information such as the random inspection date, random inspection location, number of random inspection couriers, number of random inspection pass, number of random inspection failures, random inspection pass rate, etc. The random inspection report can be displayed in the form of tables, charts, etc., so as to more intuitively understand the random inspection situation.

[0100] S502: Identify couriers who have failed the random inspection from the random inspection report, and obtain communication information of the couriers who have failed the random inspection.

[0101] In this embodiment, a list of couriers who failed the random inspection is screened out from the random inspection report. According to the list of couriers, the communication information of these couriers, such as mobile phone numbers, e-mail addresses, etc., is obtained from the database.

[0102] S503: calling a preset notification template, filling the courier's name into the preset notification template, generating notification information, and sending the notification information to the courier based on the acquired communication information.

[0103] In this embodiment, a notification template is designed and preset, which should include information such as the courier's name, inspection results, reasons for failure (such as not wearing a helmet, improper wearing of a helmet, etc.), and rectification requirements.

[0104] For each courier who fails the random inspection, fill in his name into the preset notification template to generate personalized notification information. Then, according to the courier's communication information obtained, select the appropriate communication method (such as SMS, email, etc.) to send the notification information to the corresponding courier. After sending the notification information, record the sending status for subsequent tracking and confirmation of whether the courier has received the notification.

[0105] In this embodiment, closed-loop management of the sampling inspection process is achieved by automatically counting the sampling inspection results, generating sampling inspection reports, identifying couriers who have failed, and sending notification information, thereby improving management efficiency and accuracy and ensuring the effective implementation of the regulations on couriers wearing helmets.

[0106] See also Figure 6 A sixth embodiment of a courier helmet wearing management method in an embodiment of the present invention includes:

[0107] S601. Collect helmet wearing complaint forms, and select helmet wearing complaint forms that have been reviewed according to their status.

[0108] In this embodiment, a collection mechanism for helmet wearing complaint forms is pre-established. This can be achieved through an online platform (such as a company's internal system, email, etc.). The helmet wearing complaint form includes key information such as the courier's name, work number, complaint content, complaint time, and complaint status.

[0109] S602. According to the review result of the helmet wearing complaint form, the couriers in the reviewed helmet wearing complaint form are divided into two categories: passed and failed.

[0110] In this embodiment, according to the status field of the helmet wearing complaint form, the complaint forms that have been reviewed are screened out. These complaint forms usually have a clear review result, such as "passed", "failed" or "pending review". And those couriers whose complaint result is "failed" are classified as failed review.

[0111] S603. Create a key focus list, and add the couriers whose appeals are not approved to the pre-created key focus list.

[0112] In this embodiment, a focus list is created to record the couriers whose appeals are not approved. This list can be a spreadsheet, a database, or any other form of data storage structure. The information of the couriers whose appeals are not approved (such as name, work number, etc.) is added to this list for subsequent focus and management.

[0113] In this embodiment, by collecting and analyzing helmet wearing complaint forms, couriers whose complaints have not been approved are accurately screened out and included in a pre-created key focus list, thereby effectively strengthening the supervision of couriers' safe wearing of helmets and improving safety management efficiency.

[0114] The above describes the courier helmet wearing management method in the embodiment of the present invention. The following describes the device in the embodiment of the present invention. Figure 7 , the implementation of the courier helmet wearing management device in the embodiment of the present invention includes:

[0115] The naming module 701 is used to collect real-time images of the entrance and exit of the express delivery station, and name the real-time images in the form of a name combined with a serial number after performing face recognition on the collected real-time images;

[0116] A saving module 702 is used to classify the named images according to the names of the couriers and save the classified images in a database to be detected;

[0117] The detection module 703 is used to obtain a random inspection list, obtain images corresponding to the couriers in the random inspection list from the database to be detected as random inspection images, and perform random inspection on the random inspection images to obtain random inspection results;

[0118] A statistics module 704 is used to select couriers who have failed the random inspection from the random inspection results and generate notification information corresponding to the couriers;

[0119] The management module 705 is used to collect helmet wearing complaint forms, and screen out couriers whose complaints have not been approved from the review results of the helmet wearing complaint forms, and put the couriers whose complaints have not been approved into a pre-created key focus list.

[0120] In this embodiment, the naming module 701 includes: a preprocessing unit 7011, which collects real-time images of the entrance and exit of the express station and preprocesses the collected real-time images; a first recognition unit 7012, which uses a face recognition algorithm to perform face recognition on the preprocessed real-time images, and eliminates images in which no face is recognized or the facial features are unclear, to obtain images to be named; a naming unit 7013, which is used to name the image to be named with a name combined with a serial number according to the face recognition result.

[0121] In this embodiment, the saving module 702 includes: a first classification unit 7021, which is used to classify the named images according to the couriers' names using regular expressions; a first creation unit 7022, which is used to create a courier image folder corresponding to the courier's name in the database to be detected; and a saving unit 7023, which is used to save the classified images to the corresponding courier image folder according to the classification results.

[0122] In this embodiment, the detection module 703 includes: an acquisition unit 7031, which is used to obtain a random inspection list, and use a keyword matching algorithm to obtain images corresponding to the couriers in the random inspection list from the database to be detected as random inspection images; a detection unit 7032, which is used to input the random inspection image into the improved detection model for each of the random inspection images, and obtain the predicted helmet position output by the improved detection model; a judgment unit 7033, which is used to judge whether the courier is wearing the helmet correctly according to the predicted helmet position. When it is judged that the courier is wearing the helmet correctly, the random inspection result is output as the courier is wearing the helmet correctly. Otherwise, the random inspection result is output as the courier is not wearing the helmet correctly.

[0123] In this embodiment, the statistical module 704 includes: a statistical unit 7041, which is used to count the inspection results, calculate the inspection pass rate, and generate an inspection report based on the statistical results and the inspection pass rate; a second identification unit 7042, which is used to identify couriers who failed the inspection from the inspection report, and obtain the communication information of the couriers who failed the inspection; a notification unit 7043, which is used to call a preset notification template, fill in the courier's name into the preset notification template, generate notification information, and send the notification information to the courier based on the obtained communication information.

[0124] In this embodiment, the management module 705 includes: a screening unit 7051, which is used to collect helmet wearing complaint forms, and screen out the helmet wearing complaint forms that have been reviewed according to the status of the helmet wearing complaint forms; a second classification unit 7052, which is used to classify the couriers in the helmet wearing complaint forms that have been reviewed into two categories: passed and failed according to the review results of the helmet wearing complaint forms; a second creation unit 7053, which is used to create a key focus list, and put the couriers whose complaints have not been passed into the key focus list.

[0125] In this embodiment, through real-time image collection and face recognition technology, the courier's identity and image naming are automatically performed, and the named images are classified according to the courier's name and saved in the database to be detected. When random inspection is carried out, the images corresponding to the couriers in the random inspection list are obtained from the database to be detected for random inspection according to the random inspection list, and after the random inspection is completed, the random inspection results are counted, so as to realize real-time monitoring and analysis of the couriers' helmet wearing situation. For the couriers who fail the random inspection, notification information is generated, and helmet wearing complaint forms are collected, and the couriers whose appeals are not passed are screened out from the review results of the helmet wearing complaint forms, and the couriers whose appeals are not passed are included in the pre-created key focus list. The couriers on the key focus list can be managed more strictly, which is conducive to improving the safety protection level of delivery personnel and reducing the injuries caused by traffic accidents.

[0126] Figure 7The structure of the courier helmet wearing management device shown does not constitute a limitation of the courier helmet wearing management device, and can implement the steps of the courier helmet wearing management method provided in the above-mentioned method embodiments.

[0127] above Figure 7 The courier helmet wearing management device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The courier helmet wearing management device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0128] Figure 8 It is a structural diagram of a courier helmet wearing management device provided by an embodiment of the present invention. The device 800 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 810 (for example, one or more processors) and a memory 820, and one or more storage media 830 (for example, one or more massive storage devices) storing applications 833 or data 832. Among them, the memory 820 and the storage medium 830 can be short-term storage or persistent storage. The program stored in the storage medium 830 may include one or more modules (not shown), and each module may include a series of instruction operations on the device 800. Furthermore, the processor 810 can be configured to communicate with the storage medium 830 to execute a series of instruction operations in the storage medium on the device 800.

[0129] The device 800 may also include one or more power supplies 840, one or more wired or wireless network interfaces 850, one or more input and output interfaces 860, and / or one or more operating systems 831, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc.

[0130] An embodiment of the present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the courier helmet wearing management method.

[0131] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0132] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable 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 method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0133] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for managing the wearing of helmets by couriers, characterized in that: The courier helmet wearing management method comprises: Collect real-time images of the entrances and exits of express delivery stations, perform face recognition on the collected real-time images, and name the real-time images in the form of a name combined with a serial number; Classify the named images according to the couriers’ names, and save the classified images in the database to be detected; Obtain a random inspection list, and obtain images corresponding to the couriers in the random inspection list from the database to be inspected as random inspection images, and perform random inspection on the random inspection images to obtain random inspection results; Selecting couriers who have failed the random inspection from the random inspection results, and generating notification information corresponding to the couriers; Collect helmet wearing complaint forms, and screen out couriers whose appeals have not been approved from the review results of the helmet wearing complaint forms, and put the couriers whose appeals have not been approved into a pre-created key focus list.

2. The courier helmet wearing management method according to claim 1, characterized in that: The real-time images of the express delivery station entrances and exits are collected, and the collected real-time images are named by combining the name with the serial number after face recognition, including: Collect real-time images of the entrances and exits of express delivery stations and pre-process the collected real-time images; The face recognition algorithm is used to perform face recognition on the preprocessed real-time image, and the images in which the face is not recognized or the face features are unclear are eliminated to obtain the image to be named; The image to be named is named by combining the name with the serial number according to the face recognition result.

3. The courier helmet wearing management method according to claim 1, characterized in that: The process of classifying the named images according to the names of the couriers and storing the classified images in a database to be detected includes: Use regular expressions to classify the named images according to the couriers’ names; Create a courier image folder corresponding to the courier's name in the database to be detected; According to the classification results, the classified images are saved in the corresponding courier image folder.

4. The courier helmet wearing management method according to claim 1, characterized in that: The obtaining of the random inspection list, obtaining images corresponding to the couriers in the random inspection list from the database to be inspected as random inspection images, and performing random inspection on the random inspection images to obtain random inspection results, includes: Obtain a random inspection list, and use a keyword matching algorithm to obtain images corresponding to the couriers in the random inspection list from the database to be inspected as random inspection images; For each of the sample inspection images, input the sample inspection image into the improved detection model, and obtain the predicted position of the helmet output by the improved detection model; It is determined whether the courier is wearing the helmet correctly according to the predicted position of the helmet. If so, the sampling result is outputted that the courier is wearing the helmet correctly. Otherwise, the sampling result is outputted that the courier is not wearing the helmet correctly.

5. The courier helmet wearing management method according to claim 4, characterized in that: For each of the sampled images, the sampled images are input into the improved detection model, and the predicted position of the helmet output by the improved detection model is obtained, which also includes: Build the basic structure of the YOLOv model, which includes the backbone network, feature detection layer, rough position prediction layer, and precise position prediction layer; Adding CBAM attention modules to the backbone network and the feature detection layer respectively to obtain an updated YOLOv model, modifying the anchor box calculation algorithm of the updated YOLOv model to a differential evolution algorithm, and introducing a mutation and selection mechanism in a genetic algorithm to optimize the size and shape of the anchor box, thereby obtaining an improved YOLOv model; Image data containing couriers wearing helmets are collected and annotated as a training set, and the improved YOLOv model is trained and optimized using the training set to obtain an improved detection model.

6. The courier helmet wearing management method according to claim 1, characterized in that: The method of screening out couriers who have failed the random inspection from the random inspection results and generating notification information corresponding to the couriers includes: The sampling inspection results are counted, and the sampling inspection pass rate is calculated, and a sampling inspection report is generated according to the statistical results and the sampling inspection pass rate; Identify the couriers who failed the random inspection from the random inspection report, and obtain the communication information of the couriers who failed the random inspection; The preset notification template is called, the courier's name is filled into the preset notification template, notification information is generated, and the notification information is sent to the courier based on the acquired communication information.

7. The courier helmet wearing management method according to claim 1, characterized in that: The collecting of helmet wearing complaint forms, and screening out couriers whose complaints have not been approved from the review results of the helmet wearing complaint forms, and placing the couriers whose complaints have not been approved in a pre-created key focus list, includes: Collect helmet wearing complaint forms, and filter out the reviewed helmet wearing complaint forms according to their status; According to the review results of the helmet wearing complaint form, the couriers in the helmet wearing complaint form that have been reviewed are divided into two categories: those who passed the review and those who failed the review; Create a key focus list and put the couriers whose appeals are not approved on the key focus list.

8. A courier helmet wearing management device, characterized in that: include: The naming module is used to collect real-time images of the entrances and exits of express delivery stations, and name the real-time images by combining names with serial numbers after performing face recognition on the collected real-time images; A saving module is used to classify the named images according to the names of the couriers and save the classified images in the database to be detected; A detection module is used to obtain a random inspection list, obtain images corresponding to couriers in the random inspection list from the database to be detected as random inspection images, and perform random inspection on the random inspection images to obtain random inspection results; A statistics module, used for screening out couriers who have failed the random inspection from the random inspection results, and generating notification information corresponding to the couriers; The management module is used to collect helmet wearing complaint forms, and screen out couriers whose complaints have not been approved from the review results of the helmet wearing complaint forms, and put the couriers whose complaints have not been approved into a pre-created key focus list.

9. A courier helmet wearing management device, characterized in that: comprising a memory and at least one processor, wherein the memory has computer-readable instructions stored therein; The at least one processor calls the computer-readable instructions in the memory to execute the various steps of the courier helmet wearing management method as described in any one of claims 1-7.

10. A computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by the processor, the various steps of the courier helmet wearing management method as described in any one of claims 1-7 are implemented.

Images