Farmers' market transaction behavior identification method and system combined with intelligent electronic scale

Abstract

The present invention discloses a method and system for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale, relating to the technical field of market transaction management, including: determining a first merchant area in the farmers' market, recording a continuous weighing data sequence using an intelligent electronic scale, and simultaneously tracking merchants using an industrial follow-up camera to generate an image monitoring sequence; locating a stable weighing time zone and calculating a stable weight to generate an accurate stable weighing result; performing analysis based on the image monitoring sequence to generate a first abnormal behavior identification result; comparing the stable weighing result with the payment result to identify transaction anomalies and generate a second abnormal behavior identification result; combining the two identification results to perform double abnormality verification, and triggering a transaction abnormality alarm based on the verification result. The present invention solves the technical problems of the existing technology that are unable to comprehensively monitor merchant transaction behavior and lack the ability to quickly identify and process abnormal transactions, thereby achieving the technical effect of improving the transparency and security of the farmers' market transaction process.

Description

Technical Field

[0001] The present invention relates to the technical field of market transaction management, and in particular to a method and system for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale. Background Art

[0002] With the expansion of farmers' markets and the increasing complexity of transactions, transparency and security in the transaction process have become particularly important. However, traditional farmers' market transactions still rely primarily on simple electronic scales to record weighing data, making it difficult to fully monitor merchants' operations and transaction processes. This is especially true after weighing, where merchants may replace or remove items. There is also a lack of intuitive verification of whether the user's payment amount matches the actual weighing result, which can easily lead to transaction disputes. Traditional transaction monitoring methods are unable to fully monitor merchant transactions and lack the ability to quickly identify and handle abnormal transactions. This makes it difficult to meet the needs of modern farmers' markets for efficient management and real-time monitoring of transaction behavior. Summary of the Invention

[0003] This application provides a method and system for identifying transaction behavior in farmers' markets in combination with smart electronic scales, which is used to solve the technical problems that existing technologies are unable to comprehensively monitor merchant transaction behavior and lack the ability to quickly identify and process abnormal transactions.

[0004] In view of the above problems, the present application provides a method and system for identifying transaction behaviors in farmers' markets in combination with smart electronic scales.

[0005] The first aspect of the present application provides a method for identifying transaction behavior in a farmers' market in combination with a smart electronic scale, the method comprising:

[0006] A first area of ​​a first merchant in a farmers' market is determined, wherein an intelligent electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring; when the intelligent electronic scale is activated, a continuous weighing data sequence is recorded, and at the same time, the first merchant is tracked by the industrial follow-up camera to obtain an image monitoring sequence; based on the image monitoring sequence, the weighing behavior of the first merchant is identified, and a stable weighing time zone is located; according to the stable weighing time zone, a stable weight is calculated for the continuous weighing data sequence to generate a stable weighing result; based on the image monitoring sequence, the transaction behavior between the first merchant and the user is identified, and a first abnormal behavior identification result is generated; the transaction payment platform of the first merchant is connected to obtain a payment result, and transaction anomaly identification is performed based on the stable weighing result and the payment result to generate a second abnormal behavior identification result; double abnormality verification is performed using the first abnormal behavior identification result and the second abnormal behavior identification result, and a transaction anomaly alarm is issued according to the verification result.

[0007] A second aspect of the present application provides a farmer's market transaction behavior recognition system combined with a smart electronic scale, the system comprising:

[0008] An area determination module is provided, wherein the area determination module is used to determine a first area of ​​a first merchant in the farmer's market, wherein an intelligent electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring; an image monitoring module is provided, wherein the image monitoring module is used to record a continuous weighing data sequence when the intelligent electronic scale is activated, and simultaneously track the first merchant through the industrial follow-up camera to obtain an image monitoring sequence; a weighing behavior recognition module is used to identify the weighing behavior of the first merchant based on the image monitoring sequence and locate the weighing stable time zone; a stable weight calculation module is used to calculate the continuous weighing data according to the weighing stable time zone. A weighing data sequence is used to perform stable weight calculation to generate a stable weighing result; a first transaction behavior identification module is used to identify the transaction behavior between the first merchant and the user based on the image monitoring sequence to generate a first abnormal behavior identification result; a second transaction behavior identification module is used to connect to the transaction payment platform of the first merchant to obtain the payment result, perform transaction anomaly identification based on the stable weighing result and the payment result, and generate a second abnormal behavior identification result; a double abnormality verification module is used to perform double abnormality verification with the first abnormal behavior identification result and the second abnormal behavior identification result, and issue a transaction abnormality alarm according to the verification result.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application determines a first area of ​​a first merchant in a farmers' market, wherein a smart electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring; when the smart electronic scale is activated, a continuous weighing data sequence is recorded, and at the same time, the first merchant is tracked by the industrial follow-up camera to obtain an image monitoring sequence; based on the image monitoring sequence, the weighing behavior of the first merchant is identified, and a stable weighing time zone is located; according to the stable weighing time zone, a stable weight calculation is performed on the continuous weighing data sequence to generate a stable weighing result; based on the image monitoring sequence, the transaction behavior between the first merchant and the user is identified, and a first abnormal behavior identification result is generated; the transaction payment platform of the first merchant is connected to obtain a payment result, and a transaction anomaly is identified based on the stable weighing result and the payment result to generate a second abnormal behavior identification result; double abnormality verification is performed with the first abnormal behavior identification result and the second abnormal behavior identification result, and a transaction anomaly alarm is issued according to the verification result. The present invention solves the technical problems that the existing technology cannot comprehensively monitor merchant transaction behaviors and lacks the ability to quickly identify and process abnormal transactions. It combines intelligent electronic scales and industrial follow-up cameras, and realizes accurate identification of merchant weighing and transaction behaviors through weighing data and image monitoring sequence analysis. Through stable weight calculation, payment result comparison, double abnormality verification and other technologies, it quickly discovers transaction anomalies and alarms, achieving the technical effect of improving the transparency and security of the transaction process in farmers' markets. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A flow chart of a method for identifying transaction behavior in a farmers' market in combination with a smart electronic scale provided in an embodiment of the present application;

[0013] Figure 2 Schematic diagram of the structure of the farmers' market transaction behavior identification system combined with a smart electronic scale provided in an embodiment of the present application.

[0014] Explanation of the accompanying drawings: area determination module 11, image monitoring module 12, weighing behavior recognition module 13, stable weight calculation module 14, first transaction behavior recognition module 15, second transaction behavior recognition module 16, double abnormality verification module 17. DETAILED DESCRIPTION

[0015] This application provides a method and system for identifying transaction behaviors in farmers' markets in combination with smart electronic scales, aiming to solve the technical problems that existing technologies are unable to comprehensively monitor merchant transaction behaviors and lack the ability to quickly identify and process abnormal transactions. By combining smart electronic scales and industrial follow-up cameras, accurate identification of merchant weighing behaviors and transaction behaviors is achieved through weighing data and image monitoring sequence analysis. Through stable weight calculation, payment result comparison, double abnormality verification and other technologies, transaction anomalies can be quickly discovered and alarms can be issued, achieving the technical effect of improving the transparency and security of the transaction process in farmers' markets.

[0016] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0017] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0018] Example 1, as Figure 1 As shown, the present application provides a method for identifying transaction behavior in a farmers' market in combination with a smart electronic scale, the method comprising:

[0019] Step S100: determining a first area of ​​a first merchant in a farmer's market, wherein a smart electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring.

[0020] In the embodiments of the present application, when data collection is involved, such as obtaining data through electronic scales and cameras, user permission or consent is required, and the collection, use, and processing of relevant data comply with the relevant laws, regulations, and standards of the relevant countries and regions, and will not violate the public interest or infringe on personal privacy. In monitoring trading behavior in a farmers' market, the first area of ​​the first merchant is first determined. The first area refers to the monitoring range delineated around the stall of the first merchant, which is used to centrally manage weighing and trading behaviors. This area is pre-set manually.

[0021] The first area is equipped with smart electronic scales and industrial tracking cameras. The smart electronic scales collect weighing data from merchants. The industrial tracking cameras, with high resolution and dynamic tracking capabilities, use built-in target recognition algorithms to lock onto and follow the movements of the first merchant in real time. By capturing the merchant's movements, the cameras generate complete image sequence data for behavior recognition and verification.

[0022] Step S200: When the smart electronic scale is activated, a continuous weighing data sequence is recorded, and at the same time, the first merchant is tracked by the industrial follow-up camera to obtain an image monitoring sequence.

[0023] In this embodiment of the present application, when the smart electronic scale is activated, both weight data collection and merchant behavior monitoring are initiated. The smart electronic scale records a continuous sequence of weighing data in real time, capturing weight changes. Simultaneously, an industrial tracking camera uses a pre-trained object detection model to track the first merchant and capture images of their behavior.

[0024] Specifically, the industrial tracking camera uses a target detection model to identify and lock onto the merchant's dynamic features, ensuring accurate tracking. By analyzing the merchant's displacement characteristics in consecutive image frames, it predicts the merchant's target position and dynamically adjusts the camera's control parameters (such as focal length and angle) to keep the merchant within the monitoring range. Ultimately, a complete image sequence of the merchant's behavior is recorded to generate an image monitoring sequence.

[0025] Furthermore, in the method provided in the embodiment of the application, the first merchant is tracked by the industrial follow-up camera to obtain an image monitoring sequence, and the method further includes:

[0026] A first target detection model is configured for the first merchant; images are captured in the first area using the industrial follow-up camera, and the first target detection model is called to perform target detection, thereby establishing a first image set and first camera control parameters; inter-frame dynamic features of the first merchant are analyzed based on the first image set, target displacement is predicted, and predicted displacement is generated; with the predicted displacement as a reference, the first camera control parameters are adjusted to complete tracking of the first merchant and generate the image monitoring sequence.

[0027] In this embodiment, a first object detection model is first configured. This model is trained based on labeled data of the first merchant's appearance and behavioral characteristics obtained from a historical database. The model uses the YOLO object detection algorithm to enable the model to accurately identify the first merchant's dynamic characteristics. After training, the first object detection model is deployed in the camera system for real-time detection of the first merchant.

[0028] Within the first area, the industrial tracking camera begins wide-area image acquisition, performing target detection by calling the first target detection model frame by frame. Since the initial image frame may not necessarily contain the first merchant, successive image frames are analyzed until the merchant target is successfully detected. After detecting the first merchant, its initial position is recorded and marked as the core target for subsequent tracking. Subsequently, the industrial tracking camera continues to acquire image frames containing the first merchant, forming a first image set. This image set stores the merchant's continuous dynamic features and establishes the first camera control parameters, including focal length, rotation angle, and viewing angle range, for real-time adjustment of the camera to accommodate the merchant's movements.

[0029] Next, the inter-frame dynamic features of the first merchant are analyzed based on the first image set. This process is accomplished using the optical flow method, which calculates the motion vectors of the merchant's key points in adjacent image frames, extracts their movement direction, speed, and position information, and clusters the dynamic information of these key points to form the merchant's overall motion trajectory. Subsequently, the Kalman filter algorithm is used to predict the merchant's future position. The Kalman filter model constructs a state vector, including the merchant's current coordinates, speed, and acceleration, to predict the merchant's likely position in the next frame. By smoothing the noise in the input data, the Kalman filter model outputs predicted displacement data.

[0030] The camera's control parameters are then dynamically adjusted based on the predicted displacement. Using the servo control system, the camera's rotation angle is adjusted in real time based on the predicted data to ensure the merchant remains centered in the surveillance image. Simultaneously, the camera's focal length is dynamically adjusted based on the distance between the merchant and the camera, ensuring a clear image and that the merchant is always presented at the appropriate scale. This real-time tracking process enables the camera to smoothly follow the merchant's movements, preventing the target from being lost or falling out of surveillance range.

[0031] As the camera continuously captures images, it integrates all captured frames into a complete image monitoring sequence. Each frame is timestamped to ensure time synchronization. The resulting image monitoring sequence comprehensively records the primary merchant's behavior, providing high-quality basic data for subsequent transaction analysis and anomaly detection.

[0032] Furthermore, in the method provided in the embodiment of the application, after generating the image monitoring sequence, the method further includes:

[0033] Identify the tracking object of any frame image in the image monitoring sequence; if the tracking object includes other persons except the first merchant, perform blurring processing on the other persons.

[0034] In this embodiment of the present application, an arbitrary frame of image is first extracted from the image monitoring sequence, and the deployed target detection model is called to detect and classify the targets in the image. The target detection model is able to identify all objects appearing in the frame, including the first merchant and other possible unrelated persons. Based on the appearance and behavioral features annotated in the training data, the model prioritizes the first merchant as the primary target and classifies other persons as "non-merchant objects."

[0035] After target recognition is complete, all detected objects in the frame are classified. The tracked object is the target tracked by the camera in the current frame, namely the first merchant. Other detected persons (such as passing customers or unrelated people in the background) are labeled as other persons and require further processing to protect privacy.

[0036] To ensure the privacy of unrelated persons is not disclosed, other detected persons are blurred. Specifically, the facial areas or the entire body contours of these persons are blurred using image processing techniques (such as Gaussian blur algorithm or pixelation).

[0037] After the blurring process is completed, the processed image frames are reinserted into the image monitoring sequence.

[0038] Step S300: Identify the weighing behavior of the first merchant based on the image monitoring sequence and locate the weighing stable time zone.

[0039] In an embodiment of the present application, based on the image monitoring sequence, by identifying the weighing behavior of the first merchant and combining the weighing cutoff behavior characteristics, the key image frame sequence from the merchant completing the item addition or adjustment operation to the final removal of the item is extracted, the corresponding acquisition time is obtained, and combined with the fluctuation trend of the weighing data, the time interval when the weighing data is stable is located, and finally the weighing stable time zone is determined.

[0040] Furthermore, in the method provided in the embodiment of the application, the weighing behavior of the first merchant is identified based on the image monitoring sequence, and the weighing stable time zone is located, further comprising:

[0041] Configure a weighing cutoff behavior feature, wherein the weighing cutoff behavior feature includes a continuous behavior feature of a time zone; identify the image monitoring sequence based on the weighing cutoff behavior feature, and locate the weighing image frame sequence corresponding to the weighing cutoff behavior feature; obtain the acquisition time corresponding to the weighing image frame sequence, and establish the weighing stable time zone.

[0042] In the embodiment of the present application, the weighing cutoff behavior characteristics are first configured, and the merchant's action characteristics during the weighing process are defined through rules. The weighing cutoff behavior characteristics include continuous behavior patterns within a time zone, such as the merchant's adjustment actions of adding or removing items multiple times during the weighing process, the state of the items when they are stationary, and the termination action of the items being removed from the electronic scale. Through manually set rules, such as the activity characteristics of the hand in the electronic scale area, the adjustment frequency, and the stationary time threshold, the weighing cutoff behavior standard is formed for the subsequent identification process.

[0043] Next, the image monitoring sequence is identified based on the weighing cutoff behavior characteristics. By analyzing the image monitoring sequence frame by frame, the merchant's movements are detected and extracted using frame difference and background modeling techniques. The frame difference method calculates the pixel differences between adjacent frames to identify dynamic areas and determine whether the merchant's hand or object is moving near the electronic scale. Background modeling constructs a dynamic background model to eliminate static background areas and retain only dynamic changes in the foreground. The motion features detected in consecutive image frames are statistically analyzed. When the merchant's adjustment action is detected to have terminated (e.g., the hand is no longer near the electronic scale area for more than 3 seconds) and there is no further movement of the object, these frames are marked as weighing end frames, and a weighing image frame sequence covering the entire weighing process is generated.

[0044] The timestamp of each frame in the weighing image frame sequence is then extracted and synchronized with the weight data from the electronic scale. A windowed sliding average is calculated for the weight data curve (with a window size of 1 second), and the weight fluctuation (such as standard deviation or range) is calculated within each time window. If the weight fluctuation is less than a set threshold (e.g., standard deviation less than 0.02 kg) for three consecutive seconds, the weight is considered stable and the time period is marked as a weight stability interval. Simultaneously, the merchant's behavior in the image frames is verified to ensure that the weight stability interval corresponds to the merchant's static behavior (e.g., a hand leaving the electronic scale area for more than three seconds and not returning).

[0045] Finally, by combining weight fluctuation analysis and static action verification, the time period when the weight data is stable and the merchant operation is completed is marked, which is the weighing stable time zone.

[0046] Step S400: performing stable weight calculation on the continuous weighing data sequence according to the weighing stable time zone to generate a stable weighing result.

[0047] In this embodiment, target data corresponding to a stable weighing time zone is extracted from a continuous weighing data sequence, i.e., a target continuous weighing data sequence. Fluctuation identification is then performed on the extracted data, analyzing the fluctuation amplitude of the weight data. A sliding window method is then used to determine the time period with the smallest fluctuation amplitude. Finally, the data within this stable time period is averaged to generate the final stable weighing result.

[0048] Furthermore, in the method provided in the embodiment of the application, performing stable weight calculation on the continuous weighing data sequence according to the weighing stable time zone to generate a stable weighing result, the method further includes:

[0049] According to the weighing stable time zone, a target continuous weighing data sequence of the corresponding time zone is extracted from the continuous weighing data sequence; fluctuation identification is performed on the target continuous weighing data sequence, and the average weighing data corresponding to the time zone with the smallest continuous fluctuation amplitude is obtained to generate the stable weighing result.

[0050] In this embodiment, a data subset corresponding to a specific time period is first extracted from the continuous weighing data sequence based on a calibrated stable weighing time zone. This subset is referred to as the target continuous weighing data sequence. The stable weighing time zone is a critical time period during which weight data fluctuations are minimal and reflect the actual weight of the item. Through timestamp synchronization, data points matching the stable time zone are intercepted from the continuous weighing data sequence to obtain the target continuous weighing data sequence.

[0051] Next, the system identifies fluctuations in the target continuous weighing data sequence. Using a sliding window approach (e.g., a window size of 1 second), the data sequence is gradually traversed. The weight fluctuation amplitude within each window is calculated and quantified using statistical indicators such as standard deviation or range. When the fluctuation amplitude of multiple consecutive windows is less than a preset threshold (e.g., a standard deviation less than 0.02 kg), the time period is determined to be the most stable weight data period.

[0052] The average weight data is then calculated for the identified period of minimum fluctuation. All weight data points within this period are summed and averaged to generate the final stable weight value. This calculated stable weight value is then output as the stable weighing result.

[0053] Step S500: Identify the transaction behavior between the first merchant and the user based on the image monitoring sequence, and generate a first abnormal behavior identification result.

[0054] In an embodiment of the present application, in order to generate the first abnormal behavior recognition result, based on the image monitoring sequence, by configuring the transaction end behavior feature and the transaction abnormality feature, combined with the end time of the weighing stable time zone, the transaction behavior between the first merchant and the user is analyzed. With the end point of the weighing stable time zone as the recognition starting point and the transaction end behavior feature as the recognition end point, the image monitoring sequence is frame segmented to generate a segmented image frame sequence. After that, the transaction abnormality feature is further used to collect abnormal samples, train and construct the first abnormality recognition network. The segmented image frame sequence is input into the network for analysis to identify potential abnormal behaviors in the transaction process, such as whether the merchant has changed the product after weighing is completed, etc., and finally generate the first abnormal behavior recognition result.

[0055] Furthermore, in the method provided in the embodiment of the application, identifying the transaction behavior between the first merchant and the user based on the image monitoring sequence and generating a first abnormal behavior identification result further includes:

[0056] Configure transaction cut-off behavior features and transaction abnormality features; use the end point corresponding to the weighing stable time zone as the identification starting point and the transaction cut-off behavior features as the identification end point to perform image frame segmentation on the image monitoring sequence to generate a segmented image frame sequence; collect abnormality recognition samples based on the transaction abnormality features, train and construct a first abnormality recognition network; input the segmented image frame sequence into the first abnormality recognition network for analysis to generate the first abnormal behavior recognition result.

[0057] In this embodiment, transaction end behavior features and transaction anomaly features are first configured. The transaction end behavior features describe the merchant's standard actions for completing a transaction, such as handing over the goods to the user after weighing or confirming the transaction completion. The transaction anomaly features are used to flag potential non-compliant behaviors, such as abnormal operations such as replacing goods after weighing, partially removing goods, or not delivering goods. By annotating historical transaction video data, key merchant behavior features are extracted to form a database of transaction end behavior features and transaction anomaly features.

[0058] The image monitoring sequence is then parsed frame by frame through time synchronization analysis, starting from the end of the weighing stability time zone. The frame sequence related to the transaction behavior is extracted, ending with the transaction completion action in the transaction end behavior feature. By combining timestamp and frame segmentation techniques, frame difference analysis and dynamic target detection are used to isolate the interaction action frames between the merchant and the user, generating a segmented image frame sequence covering the entire transaction process.

[0059] Based on transaction anomaly characteristics, anomaly identification samples are collected and the first anomaly identification network is trained. By simulating abnormal transaction scenarios (such as replacing items after weighing or removing parts of items), a set of labeled abnormal behavior samples is generated. Motion feature extraction techniques (such as HOG features and key point detection) are used to analyze the merchant's actions. These features are then fed into a classification model (such as SVM, random forest, or shallow neural network) for training. The trained first anomaly identification network focuses on detecting abnormal merchant actions and is capable of identifying complex abnormal behaviors.

[0060] Finally, the segmented image frame sequence is fed frame by frame into the first anomaly recognition network for analysis. The network extracts the merchant's action features and compares them with a database of anomaly features to detect any abnormal behavior. For example, it can detect whether the merchant has changed or partially removed items. Any detected abnormal behavior is labeled with its type and the corresponding timestamp and image frame information is recorded. Ultimately, the network generates a first abnormal behavior recognition result, including a description of the abnormal behavior, the time of occurrence, and the corresponding image evidence.

[0061] Furthermore, the method provided in the application embodiment also includes:

[0062] The segmented image frame sequence is identified to locate the weighing target features and the set of occluded image frames in which the weighing target features are occluded; the temporal continuity between the image frames in the occluded image frame set is analyzed to identify continuous occlusion intervals; a reliability analysis is performed on the first abnormal behavior recognition result based on the continuous occlusion intervals to generate a first reliability index; and the first abnormal behavior recognition result is optimized based on the first reliability index.

[0063] In an embodiment of the present application, the segmented image frame sequence is first identified to locate the weighing target features. Specifically, a target detection algorithm (such as YOLOv4) is used to detect the weighing commodity features in each frame of the image through a trained model. The training process of the model is to first extract the image data of the weighing commodity from the historical monitoring video, manually mark the bounding box and category of the commodity, and form a training data set. Through multiple rounds of iterative training, the performance of the model's bounding box regression, classification, and confidence prediction is optimized. After training, the model can generate the bounding box, detection confidence, and position features of the weighing commodity in each frame of the image. The monitoring confidence is a quantitative value of the model's credibility of the target detection result (ranging from 0 to 1). Through detection, a feature set of the weighing target in each frame of the image is obtained, including position, size, and confidence.

[0064] Next, the occlusion of the weighing target in the image is detected, and a set of occluded image frames is generated. Specifically, occlusion judgment is made based on the confidence of target detection and the regional integrity of the bounding box. When the confidence of a frame is lower than a preset threshold (such as 0.5), the frame is marked as a possible occlusion frame; if the area of ​​the bounding box is reduced by more than 50% compared to the previous frame, it is further confirmed that the target feature is partially occluded. All image frames that meet the occlusion conditions are classified into the occluded image frame set, and their corresponding timestamps are recorded. Through this step, a set containing all occluded frames is generated, and the occlusion information of each frame is annotated.

[0065] The timestamps in the occluded image frames are then analyzed to identify consecutive intervals of occlusion. Using the timestamp interval method, the occluded image frames are arranged in chronological order and the time interval between adjacent occluded frames is calculated. If the interval is less than a set value (e.g., 1 second), these frames are considered to be continuously occluded. The start, end, and duration of each occlusion are calculated, and the time range and length of each occlusion are marked. This step identifies all consecutive intervals of occlusion.

[0066] Next, a reliability analysis is performed on the first abnormal behavior recognition result based on the intervals between consecutive occlusions to generate a first reliability index. This reliability analysis considers the duration of the occlusion and the degree of target feature loss. The proportion of the continuous occlusion time to the total segmented image frame sequence time, denoted as P, is calculated, and the degree of feature loss in the occluded frames (e.g., the proportion of the bounding box area reduction), denoted as F, is calculated. The difference between 1 minus P and 1 minus F is calculated, and these two differences are weighted, with each difference having equal weight. The first reliability index (ranging from 0 to 1) is obtained through this calculation.

[0067] Finally, the first reliability index is used to optimize the abnormal behavior recognition results. If the reliability is low (e.g., the first reliability index is below 0.6), a re-detection process is triggered, ignoring occluded frames and re-detecting the completeness and consistency of the key abnormal behavior actions in only the unoccluded frames. Specifically, all frames marked as occluded are first removed from the segmented image frame sequence, retaining only the unoccluded frames for subsequent analysis. Then, in the unoccluded frames, the key abnormal behavior actions are re-detected, such as whether the merchant has replaced or partially removed an item. Motion analysis algorithms (such as frame difference or keypoint tracking) are used to extract the merchant's hand movements and the trajectory of the weighing target to verify the occurrence of abnormal behavior. Finally, the motion feature sequence from the unoccluded frames is combined to verify the completeness and consistency of the abnormal behavior features. For example, the process of weighing target changes is checked to ensure logical consistency (e.g., removing an item corresponds to a decrease in mass), and the time series of abnormal features is verified to be continuous. If the abnormal feature sequence can be reasonably compensated in the unoccluded frames, it is considered complete. If the key feature loss caused by occlusion cannot be compensated, it is marked as a false positive and the erroneous result is eliminated.

[0068] For results with a reliability higher than a set threshold (e.g., 0.6), the abnormal behavior recognition result is directly confirmed and the reliability score is added as output information. The optimized results include the type of abnormal behavior, the time of occurrence, the relevant image frames, and their reliability indicators.

[0069] Step S600: Connecting to the transaction payment platform of the first merchant to obtain a payment result, performing transaction anomaly identification based on the stable weighing result and the payment result, and generating a second abnormal behavior identification result.

[0070] In this embodiment of the present application, the system first connects to the first merchant's transaction payment platform to obtain the payment result. The system then compares the stable weighing result with the payment result to generate a second abnormal behavior identification result. The transaction payment platform is the system through which merchants and users complete payment transactions and records key information such as the unit price, payment amount, and payment weight of the goods. A secure API interface is used to establish real-time data connection with the payment platform, extracting the merchant's transaction records from the payment platform, including product information, payment amount, and payment weight.

[0071] After obtaining the payment result, the transaction weight corresponding to the payment result is calculated using the payment amount and the unit price of the product. The transaction weight is calculated by dividing the unit price of the product by the payment amount.

[0072] The calculated transaction weight is then directly compared with the stable weighing result to determine their consistency. If the deviation exceeds a set tolerance threshold (e.g., 3%), the transaction is flagged as abnormal and a second abnormal behavior identification result is generated. This second abnormal behavior identification result is used to determine whether the merchant calculated fees based on the stable weighing result, identifying possible abnormal behavior. For example, if the payment result corresponds to a weight higher than the stable weighing result, it is marked as "overcharged," indicating that the merchant is overcharging or engaging in malicious behavior.

[0073] Step S700: performing double abnormality verification based on the first abnormal behavior recognition result and the second abnormal behavior recognition result, and issuing a transaction abnormality alarm based on the verification results.

[0074] In this embodiment of the present application, when dual anomaly verification is performed based on both the first and second abnormal behavior identification results, a transaction anomaly alarm is triggered if either result indicates an anomaly. Specifically, if the first abnormal behavior identification result indicates that the merchant performed an abnormal operation after weighing (such as replacing or removing an item), or if the second abnormal behavior identification result indicates that the payment behavior does not match the weighing data (such as overcharging), the transaction will be directly determined to be abnormal, without requiring both abnormalities. During verification, time matching and behavioral logic checks are combined to ensure accurate anomaly detection.

[0075] Once an anomaly is confirmed, whether it's related to weighing or payment, a transaction anomaly alert is immediately generated. Alert information includes the anomaly type (e.g., weight change, item removal, overcharge), the timeframe in which the anomaly occurred, and the associated weight and fee deviation data. This mechanism enables rapid identification and alerting, ensuring transparency and compliance with transaction regulations.

[0076] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0077] This application determines a first area of ​​a first merchant in a farmers' market, wherein a smart electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring; when the smart electronic scale is activated, a continuous weighing data sequence is recorded, and at the same time, the first merchant is tracked by the industrial follow-up camera to obtain an image monitoring sequence; based on the image monitoring sequence, the weighing behavior of the first merchant is identified, and a stable weighing time zone is located; according to the stable weighing time zone, a stable weight calculation is performed on the continuous weighing data sequence to generate a stable weighing result; based on the image monitoring sequence, the transaction behavior between the first merchant and the user is identified, and a first abnormal behavior identification result is generated; the transaction payment platform of the first merchant is connected to obtain a payment result, and a transaction anomaly is identified based on the stable weighing result and the payment result to generate a second abnormal behavior identification result; double abnormality verification is performed with the first abnormal behavior identification result and the second abnormal behavior identification result, and a transaction anomaly alarm is issued according to the verification result. The present invention solves the technical problems that the existing technology cannot comprehensively monitor merchant transaction behaviors and lacks the ability to quickly identify and process abnormal transactions. It combines intelligent electronic scales and industrial follow-up cameras, and realizes accurate identification of merchant weighing and transaction behaviors through weighing data and image monitoring sequence analysis. Through stable weight calculation, payment result comparison, double abnormality verification and other technologies, it quickly discovers transaction anomalies and alarms, achieving the technical effect of improving the transparency and security of the transaction process in farmers' markets.

[0078] Example 2 is based on the same inventive concept as the method for identifying transaction behavior in farmers' markets in combination with smart electronic scales in the above embodiment. Figure 2 As shown, the present application provides a farmer's market transaction behavior identification system combined with a smart electronic scale. The system and method embodiments in the present application are based on the same inventive concept. The system includes:

[0079] An area determination module 11 is used to determine a first area of ​​a first merchant in a farmer's market, wherein an intelligent electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring; an image monitoring module 12 is used to record a continuous weighing data sequence when the intelligent electronic scale is activated, and simultaneously track the first merchant through the industrial follow-up camera to obtain an image monitoring sequence; a weighing behavior recognition module 13 is used to identify the weighing behavior of the first merchant based on the image monitoring sequence and locate the weighing stable time zone; a stable weight calculation module 14 is used to calculate the weighing behavior of the first merchant according to the weighing stable time zone. A continuous weighing data sequence is used to perform stable weight calculation to generate a stable weighing result; a first transaction behavior identification module 15, the first transaction behavior identification module 15 is used to identify the transaction behavior between the first merchant and the user based on the image monitoring sequence, and generate a first abnormal behavior identification result; a second transaction behavior identification module 16, the second transaction behavior identification module 16 is used to connect to the transaction payment platform of the first merchant to obtain the payment result, perform transaction anomaly identification based on the stable weighing result and the payment result, and generate a second abnormal behavior identification result; a double abnormality verification module 17, the double abnormality verification module 17 is used to perform double abnormality verification with the first abnormal behavior identification result and the second abnormal behavior identification result, and issue a transaction abnormality alarm according to the verification result.

[0080] Furthermore, the system is also used to implement the following functions:

[0081] Configure a weighing cutoff behavior feature, wherein the weighing cutoff behavior feature includes a continuous behavior feature of a time zone; identify the image monitoring sequence based on the weighing cutoff behavior feature, and locate the weighing image frame sequence corresponding to the weighing cutoff behavior feature; obtain the acquisition time corresponding to the weighing image frame sequence, and establish the weighing stable time zone.

[0082] Furthermore, the system is also used to implement the following functions:

[0083] Configure transaction cut-off behavior features and transaction abnormality features; use the end point corresponding to the weighing stable time zone as the identification starting point and the transaction cut-off behavior features as the identification end point to perform image frame segmentation on the image monitoring sequence to generate a segmented image frame sequence; collect abnormality recognition samples based on the transaction abnormality features, train and construct a first abnormality recognition network; input the segmented image frame sequence into the first abnormality recognition network for analysis to generate the first abnormal behavior recognition result.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] The segmented image frame sequence is identified to locate the weighing target features and the set of occluded image frames in which the weighing target features are occluded; the temporal continuity between the image frames in the occluded image frame set is analyzed to identify continuous occlusion intervals; a reliability analysis is performed on the first abnormal behavior recognition result based on the continuous occlusion intervals to generate a first reliability index; and the first abnormal behavior recognition result is optimized based on the first reliability index.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] A first target detection model is configured for the first merchant; images are captured in the first area using the industrial follow-up camera, and the first target detection model is called to perform target detection, thereby establishing a first image set and first camera control parameters; inter-frame dynamic features of the first merchant are analyzed based on the first image set, target displacement is predicted, and predicted displacement is generated; with the predicted displacement as a reference, the first camera control parameters are adjusted to complete tracking of the first merchant and generate the image monitoring sequence.

[0088] Furthermore, the system is also used to implement the following functions:

[0089] Identify the tracking object of any frame image in the image monitoring sequence; if the tracking object includes other persons except the first merchant, perform blurring processing on the other persons.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] According to the weighing stable time zone, a target continuous weighing data sequence of the corresponding time zone is extracted from the continuous weighing data sequence; fluctuation identification is performed on the target continuous weighing data sequence, and the average weighing data corresponding to the time zone with the smallest continuous fluctuation amplitude is obtained to generate the stable weighing result.

[0092] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0094] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for identifying transaction behaviors in farmers' markets using smart electronic scales, characterized in that: include: Determine a first area of ​​a first merchant in a farmers' market, wherein a smart electronic scale and an industrial tracking camera are provided in the first area, and the industrial tracking camera follows the first merchant for monitoring; When the smart electronic scale is activated, a continuous weighing data sequence is recorded, and at the same time, the first merchant is tracked by the industrial tracking camera to obtain an image monitoring sequence; identifying the weighing behavior of the first merchant based on the image monitoring sequence, and locating a weighing stable time zone; Calculating the stable weight of the continuous weighing data sequence according to the weighing stable time zone to generate a stable weighing result includes: According to the weighing stable time zone, extracting a target continuous weighing data sequence corresponding to the time zone from the continuous weighing data sequence; Performing fluctuation identification on the target continuous weighing data sequence, obtaining average weighing data corresponding to the time zone with the smallest continuous fluctuation amplitude, and generating the stable weighing result; identifying a transaction behavior between the first merchant and the user based on the image monitoring sequence, and generating a first abnormal behavior recognition result; Connecting to the transaction payment platform of the first merchant to obtain a payment result, performing transaction anomaly identification based on the stable weighing result and the payment result, and generating a second abnormal behavior identification result; Double abnormality verification is performed using the first abnormal behavior recognition result and the second abnormal behavior recognition result, and a transaction abnormality alarm is issued based on the verification results.

2. The method for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale according to claim 1, characterized in that: Identifying the weighing behavior of the first merchant based on the image monitoring sequence and locating a weighing stable time zone includes: Configuring a weighing cutoff behavior feature, wherein the weighing cutoff behavior feature includes a continuous behavior feature of a time zone; identifying the image monitoring sequence based on the weighing cutoff behavior feature, and locating a weighing image frame sequence corresponding to the weighing cutoff behavior feature; The acquisition time corresponding to the weighing image frame sequence is obtained to establish the weighing stable time zone.

3. The method for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale according to claim 2, wherein: Identifying the transaction behavior between the first merchant and the user based on the image monitoring sequence and generating a first abnormal behavior identification result includes: Configure transaction cut-off behavior characteristics and transaction abnormality characteristics; Taking the end time corresponding to the weighing stable time zone as the identification starting point and the transaction end behavior feature as the identification end point, performing image frame segmentation on the image monitoring sequence to generate a segmented image frame sequence; Collecting anomaly recognition samples based on the transaction anomaly features, and training and constructing a first anomaly recognition network; The segmented image frame sequence is input into the first anomaly recognition network for analysis to generate the first abnormal behavior recognition result.

4. The method for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale according to claim 3, wherein: Also includes: Identifying the segmented image frame sequence, locating weighing target features, and a set of occluded image frames in which the weighing target features are occluded; Analyzing the temporal continuity between the image frames in the occluded image frame set to identify continuous occlusion intervals; Performing a reliability analysis on the first abnormal behavior recognition result based on the continuous occlusion interval to generate a first reliability index; The first abnormal behavior recognition result is optimized based on the first reliability indicator.

5. The method for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale according to claim 1, wherein: The first merchant is tracked by the industrial follow-up camera to obtain an image monitoring sequence, including: configuring a first object detection model for the first merchant; Capturing images in the first area using the industrial follow-up camera, and calling the first target detection model to perform target detection, thereby establishing a first image set and first camera control parameters; Analyzing inter-frame dynamic features of the first merchant based on the first image set, performing target displacement prediction, and generating predicted displacement; The first camera control parameters are adjusted based on the predicted displacement to complete tracking of the first merchant and generate the image monitoring sequence.

6. The method for identifying transaction behavior in a farmers' market in combination with an intelligent electronic scale according to claim 1, wherein: After generating the image monitoring sequence, the method further includes: Identifying a tracking object in any frame image in the image monitoring sequence; If the tracking object includes other persons except the first merchant, the other persons are blurred.

7. The farmer's market transaction behavior recognition system combined with the intelligent electronic scale is characterized by: The system is used to execute the method for identifying transaction behavior in a farmers' market in combination with a smart electronic scale according to any one of claims 1 to 6, and the system comprises: an area determination module, the area determination module being configured to determine a first area of ​​a first merchant in the farmers' market, wherein a smart electronic scale and an industrial follow-up camera are provided in the first area, and the industrial follow-up camera follows the first merchant for monitoring; an image monitoring module, configured to record a continuous weighing data sequence when the intelligent electronic scale is activated, and simultaneously track the first merchant through the industrial follow-up camera to obtain an image monitoring sequence; a weighing behavior recognition module, configured to recognize the weighing behavior of the first merchant based on the image monitoring sequence and locate a stable weighing time zone; a stable weight calculation module, configured to perform stable weight calculation on the continuous weighing data sequence according to the weighing stable time zone to generate a stable weighing result; a first transaction behavior recognition module, configured to recognize the transaction behavior between the first merchant and the user based on the image monitoring sequence and generate a first abnormal behavior recognition result; a second transaction behavior identification module, the second transaction behavior identification module being configured to connect to the transaction payment platform of the first merchant to obtain a payment result, perform transaction anomaly identification based on the stable weighing result and the payment result, and generate a second abnormal behavior identification result; A double abnormality verification module is used to perform double abnormality verification based on the first abnormal behavior recognition result and the second abnormal behavior recognition result, and issue a transaction abnormality alarm based on the verification results.

Images