Dynamic monitoring system and method for business hall
Through the dynamic monitoring system of the business hall, image acquisition and human detection technology are used to monitor the staff situation in real time and deploy personnel in a timely manner, solving the problem of untimely personnel dispatch in the business hall and improving customer experience and service quality.
Patent Information
- Application Number
- CN202410989304.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
AI Technical Summary
Due to the inadequate staff dispatch in the business hall, customers wait for too long, which affects customer experience and service quality.
A dynamic monitoring system for business halls is designed, including monitoring module, image preprocessing module, detection module, area judgment module and alarm module. The video stream is obtained through the image acquisition device, the overall image is spliced, the target object is detected, and its type is judged. Personnel allocation is carried out according to preset alarm conditions.
It effectively avoids the situation where customers wait too long and personnel leave their posts, improves customer experience and service efficiency, and improves the operation management and service quality of the business hall.
Smart Images

Figure CN118945291A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a business hall dynamic monitoring system and method. Background Art
[0002] In modern society, the business hall of a power supply station, as an important window for power services, undertakes a large number of customer consultation, business processing and other service work. However, due to the work arrangement of the power supply station or other reasons, there may be a phenomenon that the front desk staff of the business hall is not in place, resulting in no relevant staff to assist customers when they go to the business hall to handle business. This situation will seriously affect the customer experience, increase the customer's waiting time, and cause dissatisfaction. In addition, during certain periods of time, there may be too many users and insufficient staff, which further aggravates the customer's waiting time and dissatisfaction.
[0003] Long waiting times and lack of effective staff assistance often lead to customers raising opinions or even complaints, which has a negative impact on the service quality of the power supply station. How to effectively monitor the staff presence in the business hall and the number of customers, and make timely staff deployment, has become an urgent problem to be solved. Summary of the invention
[0004] The present invention provides a business hall dynamic monitoring system and method to solve the problem of untimely personnel dispatch in the business hall.
[0005] According to one aspect of the present invention, a business hall dynamic monitoring system is provided, the system comprising: a monitoring module, an image preprocessing module, a detection module, a region judgment module and an alarm module;
[0006] The monitoring module is used to obtain a video stream collected by at least one image acquisition device, and to capture images of each of the video streams to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each of the video streams, and each of the image acquisition devices has different posture information, and the posture information includes at least one of a position and an angle;
[0007] The image preprocessing module is used to splice the images to be processed in each of the image data sets to obtain an overall image of the business hall;
[0008] The detection module is used to detect the overall image of each business hall and obtain a detection frame of the target object;
[0009] The area determination module is used to determine the type of each target object according to the pre-divided working area and the detection frame of the target object, and the types of the target objects at least include staff and non-staff;
[0010] The alarm module is used to determine whether a preset alarm condition is met according to the type of each target object. If so, an alarm is issued according to the preset alarm rules to deploy staff.
[0011] According to another aspect of the present invention, a method for dynamic monitoring of a business hall is provided. The method is executed by a dynamic monitoring system for a business hall described in any one of the embodiments of the present invention. The method comprises:
[0012] Acquire a video stream acquired by at least one image acquisition device, and perform image capture on each of the video streams to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each of the video streams, and each of the image acquisition devices has different posture information, and the posture information includes at least one of a position and an angle;
[0013] Splicing each frame of the image to be processed to obtain an overall image of the business hall;
[0014] Detecting the entire image of the business hall to obtain a detection frame of the target object;
[0015] Determine the type of each target object according to the pre-divided working area and the detection frame of the target object, wherein the type of the target object at least includes staff and non-staff;
[0016] It is determined whether the preset alarm conditions are met according to the type of each target object. If so, an alarm is issued according to the preset alarm rules to deploy staff.
[0017] The embodiment of the present invention provides a system and method for dynamic monitoring of a business hall, the system comprising: a monitoring module, used to obtain a video stream collected by at least one image acquisition device, and to capture images of each video stream to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each video stream, and the posture information of each image acquisition device is different, and the posture information includes at least one of a position and an angle; an image preprocessing module, used to splice the images to be processed in each image data set to obtain an overall image of the business hall; a detection module, used to detect the overall image of each business hall and obtain a detection frame of a target object; an area judgment module, used to judge the type of each target object according to a pre-divided working area and a detection frame of the target object, and if so, the type of the target object at least includes staff and non-staff; an alarm module, used to judge whether a preset alarm condition is met according to the type of each target object, and if so, to issue an alarm according to a pre-set alarm rule. An image data set is obtained by intercepting the video streams collected by image acquisition devices with different posture information, and an overall image of the business hall is obtained by splicing the images to be processed in the image data set, thereby realizing all-round monitoring of the business hall. A detection frame of the target object is obtained by detecting the overall image, and the type of the target object is judged according to the pre-divided work area to determine whether the target object is a staff member or a non-staff member. When the alarm conditions are met, an alarm is given in a timely manner according to the alarm rules, and personnel are deployed to avoid customers waiting too long when handling business and staff being away from their posts for a long time, thereby improving customer experience, improving customer service efficiency and satisfaction, and improving the operation management and service quality of the business hall.
[0018] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] 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.
[0020] Figure 1 This is a schematic diagram of the structure of a business hall dynamic monitoring system provided according to Embodiment 1 of the present invention;
[0021] Figure 2 This is a schematic diagram of the structure of a business hall dynamic monitoring system provided according to Embodiment 2 of the present invention;
[0022] Figure 3 is a flow chart of a method for dynamic monitoring of a business hall provided according to Embodiment 3 of the present invention;
[0023] Figure 4 It is a structural schematic diagram of an electronic device for implementing the business hall dynamic monitoring method according to an embodiment of the present invention. DETAILED DESCRIPTION
[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0025] It should be noted that the terms "first", "second", etc. 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 of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "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.
[0026] Embodiment 1
[0027] Figure 1 This is a schematic diagram of the structure of a business hall dynamic monitoring system provided by the first embodiment of the present invention. This embodiment is applicable to the situation of timely dispatching staff in the business hall. Figure 1 As shown, the system includes: a monitoring module 11, an image preprocessing module 12, a detection module 13, a region determination module 14 and an alarm module 15;
[0028] The monitoring module 11 is used to obtain a video stream collected by at least one image acquisition device, and to capture images of each video stream to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each video stream, and each image acquisition device has different posture information, and the posture information includes at least one of a position and an angle;
[0029] An image preprocessing module 12 is used to splice the images to be processed in each image data set to obtain an overall image of the business hall;
[0030] The detection module 13 is used to detect the overall image of each business hall and obtain the detection frame of the target object;
[0031] An area determination module 14 is used to determine the type of each target object according to the pre-divided working area and the detection frame of the target object, and the type of the target object at least includes staff and non-staff;
[0032] The alarm module 15 is used to determine whether the preset alarm conditions are met according to the type of each target object. If so, an alarm is issued according to the preset alarm rules to deploy staff.
[0033] In this embodiment, the monitoring module 11 can be understood as a data processing module for monitoring the personnel in the business hall; the image acquisition device can be a camera, a video camera, an infrared thermal imaging device, etc. The image data set can be understood as a data set formed by at least one image at the same time; the image to be processed can be understood as an image with processing requirements; the posture information can be understood as information describing the posture of the image acquisition device, and the posture information includes at least one of the position and the angle. The range of the image acquisition device can be determined according to the posture information of the image acquisition device and the parameters of the image acquisition device itself.
[0034] An image acquisition device is set in advance in the business hall. The number of image acquisition devices can be set according to the size and layout of the business hall. When the business hall is large, multiple image acquisition devices can be set to avoid monitoring blind spots and ensure that customers in the business hall and staff in working state can be monitored. When the business hall is small, a small number of image acquisition devices are set accordingly; at the same time, the number and posture of the image acquisition devices can be reasonably selected according to the layout of the business hall. The image acquisition device is set according to the actual situation of the business hall. When multiple image acquisition devices are set, the posture information of each image acquisition device is different. After each image acquisition device is installed, the video stream can be collected in real time. The monitoring module 11 obtains the video stream collected by at least one image acquisition device. The monitoring module 11 can periodically obtain a video stream at a certain interval, and the length of the video stream can also be pre-set; the image in each video stream is intercepted, and an image data set is formed based on the intercepted image to be processed at the same time. For example, the number of image acquisition devices is 5, and the image at time t1 in the video stream is intercepted to obtain 5 images at time t1, forming an image data set; the image at time t2 in the video stream is intercepted to obtain 5 images at time t2, forming an image data set..., and images at different times are intercepted in the above manner to form multiple image data sets.
[0035] In this embodiment, the image preprocessing module 12 can be understood as a data processing module that preprocesses images. Since different image acquisition devices acquire images at different locations in the business hall, which are usually only partial images of the business hall, it is necessary to process these partial images to obtain the overall image of the business hall. For each image data set, the image preprocessing module 12 splices the images to be processed in the image data set, splices the acquired images to be processed at different locations of the business hall into an overall image, displays the entirety of the business hall, and obtains the overall image of the business hall through image splicing.
[0036] In this embodiment, the detection module 13 can be understood as a data processing module for detecting staff and customers in the business hall. The target object can be understood as a detected person, which can be a staff member or a customer handling business. The detection module 13 performs human body detection on the overall image of the business hall according to a pre-set detection algorithm or a pre-trained neural network model, determines the location where the target object may exist, and obtains a detection frame of the target object.
[0037] In this embodiment, the area judgment module 14 can be understood as a data processing module for judging the area where the target object is located. The working area is pre-divided, and the area judgment module 14 judges whether the detection frame of the target object is within the working area according to the pre-divided working area, and determines the type of the target object according to the area where the target object is located. The types of target objects are divided into staff and non-staff, and the staff can also be subdivided. For example, the staff is divided into staff who handle business and staff who provide guidance, and so on. The non-staff in the embodiment of the present application are usually customers, and the non-staff can also be further refined.
[0038] In this embodiment, the alarm module 15 can be understood as a data processing module for alarming. The alarm conditions can be preset, for example, the number of staff is lower than a certain threshold, or the number of staff is lower than the threshold within a certain period of time, or the number of non-staff is higher than a certain threshold, the ratio of the number of non-staff to the number of staff is higher than a certain threshold, and so on. The alarm rules can be sending information, playing prompt sounds, and so on. The alarm module 15 analyzes the type of each target object to determine whether the preset alarm conditions are met. If so, the preset alarm rules are determined, and alarm prompts are issued according to the alarm rules so as to timely deploy the staff, for example, sending alarm information to management personnel, and the management personnel timely adjust the number of staff to provide better services, and so on.
[0039] An embodiment of the present invention provides a dynamic monitoring system for a business hall, which obtains an image data set by intercepting a video stream collected by an image acquisition device with different posture information, obtains an overall image of the business hall by splicing the images to be processed in the image data set, and realizes all-round monitoring of the business hall. By detecting the overall image, a detection frame of the target object is obtained, and the type of the target object is judged according to the pre-divided work area to determine whether the target object is a staff member or a non-staff member. When the alarm condition is met, an alarm is issued in time according to the alarm rules, and personnel are deployed to avoid the situation where customers wait too long when handling business and the staff are away from their posts for a long time, thereby improving customer experience, improving customer service efficiency and satisfaction, and improving the operation management and service quality of the business hall.
[0040] Embodiment 2
[0041] Figure 2 This is a schematic diagram of the structure of a business hall dynamic monitoring system provided in Embodiment 2 of the present invention. This embodiment is detailed based on the above embodiment.
[0042] The monitoring module 11 includes: a video stream acquisition unit 111 and an image capture unit 112;
[0043] The video stream acquisition unit 111 is used to acquire the video stream through the uniform resource locator of each image acquisition device using the real-time streaming protocol;
[0044] The image capture unit 112 is used to capture a frame of to-be-processed image at the same time in each video stream when the image capture condition is met.
[0045] In this embodiment, the real-time streaming protocol is the RTSP protocol; the uniform resource locator is the URL; the image capture condition can be pre-set, for example, the image capture period T is pre-set, and when it is detected that the time difference between the current moment and the last time an image was captured reaches the period T, it is determined that the image capture condition is met; or, when a trigger operation, control instruction or other information is received, it is determined that the image capture condition is met.
[0046] RTSP (Real-time Streaming Protocol) is a network control protocol designed for transmitting real-time data on an IP network. The video stream acquisition unit 111 uses the OpenCV library in Python to capture the video stream through the URL of the specified image acquisition device (e.g., camera), such as rtsp: / / username@camera_ip / stream. OpenCV provides the cv2.VideoCapture function for reading the video stream from the camera in real time.
[0047] The image capture conditions are set in advance, and the image capture unit 112 determines in real time whether the image capture conditions are met. When the image capture conditions are met, a frame of image at the same time in each video stream is captured, and these captured images are used as images to be processed. The images to be processed can be accumulated and stored in a database. The embodiment of the present application can capture images at regular intervals, and the timed capture process can be implemented through the time module in Python. The time interval is set using the time.sleep(interval) function to ensure that a frame of image is read from the video stream at fixed intervals. Each image read is obtained through the cv2.VideoCapture.read() function, and the cv2.imwrite function is used to save the image to the local file system or directly store it in the database. By accumulating and storing the captured images in the database, continuous monitoring of the business hall and query and analysis of historical images can be achieved, which provides basic data support for subsequent image preprocessing and human body detection.
[0048] Optionally, the image preprocessing module 12 includes an image processing unit 121 and an image stitching unit 122;
[0049] The image processing unit 121 is used to perform denoising on each image to be processed by Gaussian filtering, and to perform contrast enhancement on each denoised image by histogram equalization to obtain an enhanced image;
[0050] The image stitching unit 122 is used to perform boundary stitching and deduplication processing on each enhanced image according to the corresponding position information to obtain an overall image of the business hall.
[0051] The image processing unit 121 uses Gaussian filtering to perform denoising on each intercepted image to be processed, and the calculation formula is:
[0052]
[0053] Among them, σ is the standard deviation, which determines the size and blur degree of the filter; the cv2.GaussianBlur function in OpenCV is used to perform Gaussian filtering on the intercepted image to be processed, thereby effectively removing high-frequency noise in the image and improving image quality.
[0054] The image processing unit 121 uses histogram equalization to further enhance the contrast of each image after denoising to improve the effect of subsequent detection. The calculation formula is:
[0055]
[0056] Wherein, h(v) is the cumulative probability distribution of gray value v, f(i) is the number of pixels of gray level i, and n is the total number of pixels. Using the cv2.equalizeHist function in OpenCV to perform histogram equalization on the image can significantly improve the contrast of the image, so that the subsequent detection algorithm can more accurately identify the target object in the image. Through image denoising and contrast enhancement processing, the image preprocessing module 12 can effectively improve the image quality, provide clearer and more accurate image data for the subsequent detection module 13, and thus improve the detection accuracy and reliability of the entire system.
[0057] The image stitching unit 122 determines the posture information corresponding to each enhanced image, and the posture information can be determined according to the corresponding relationship between the image and the image acquisition device. The enhanced images are stitched at the boundaries according to the posture information, and the images are stitched into a complete large image. Since the acquisition areas of the image acquisition devices may overlap, a person may be captured by two image acquisition devices at the same time. After the stitching is completed, the image is deduplicated, and the overlapping parts are removed or fused, etc., to finally obtain the overall image of the business hall.
[0058] The detection module 13 includes: a model loading unit 131 and a detection unit 132;
[0059] A model loading unit 131 is used to load a model weight file and a configuration file of a pre-trained object detection model, and complete the construction of the object detection model based on the model weight file and the configuration file;
[0060] The detection unit 132 is used to process the overall image of the business hall through the object detection model, divide the overall image of the business hall into at least one grid, determine the detection frame and class probability predicted for each grid, and determine the detection frame of the target object based on the detection frame and class probability.
[0061] In this embodiment, the object detection model can be understood as a model for detecting a target object, which can be a neural network model, for example, a YOLO model.
[0062] The model loading unit 131 uses the cv2.dnn.readNet function in OpenCV to load the model weight file and configuration file of the object detection model, and completes the construction of the object detection model based on the model weight file and configuration file to ensure that the model is ready.
[0063] The detection unit 132 processes the overall image of the business hall through the object detection model, for example, through the convolutional neural network in the YOLO model to process the image; the overall image of the business hall is divided into one or more grids, and the detection frame and category probability predicted by each grid are determined to achieve human body detection. The detection frame where the person exists is determined based on the detection frame predicted by each grid and the category probability, so as to obtain the detection frame where the target object is located.
[0064] Taking the object detection model as the YOLO model as an example, the YOLO model divides the input image into S×S grids, each grid is responsible for detecting an object, and each grid unit predicts B detection boxes and their confidences and conditional probabilities of C categories. The specific calculations include: the position, width, height and confidence of each detection box. The confidence reflects the possibility that the detection box does contain an object and the degree of overlap between the predicted box and the actual box.
[0065] The detection result includes the coordinates of the detection box and the category probability.
[0066] The detection module 13 may further include: a hardware acceleration unit, which is used to utilize a GPU to perform calculation acceleration to improve the detection speed.
[0067] Since the YOLO model is computationally complex and requires processing a large number of convolution operations, using GPU for parallel computing can significantly improve the processing speed.
[0068] By setting the OpenCV DNN module to use the CUDA backend (cv2.dnn.DNN_BACKEND_CUDA) and CUDA target (cv2.dnn.DNN_TARGET_CUDA), the computing tasks can be assigned to the GPU for execution, greatly accelerating the model's reasoning speed and real-time detection capabilities.
[0069] Optionally, the system further includes: a region division module 16;
[0070] The area division module 16 is used to receive user operations and perform area division based on the user operations to form a work area and a customer area.
[0071] In this embodiment, the area division module 16 can be understood as a data processing module for dividing different areas. In this embodiment of the application, the area is divided into a work area and a customer area. The work area refers to the area where the staff is located during the process of handling business for the customer, and the customer area refers to the area where the customer is located. The user operation can be an operation triggered by the user clicking the mouse, the screen, etc.
[0072] The user can trigger the user operation by moving, clicking the mouse, clicking the screen, etc. The area division module 16 receives the user operation and determines the point or area selected by the user according to the user operation, thereby performing area division to form a work area and a customer area.
[0073] The area division module 16 includes: an image display unit 161, a user operation receiving unit 162, a static area configuration unit 163 and a dynamic adjustment unit 164;
[0074] An image display unit 161 is used to display the image to be divided, where the image to be divided is an image obtained by photographing the entire business hall;
[0075] A user operation receiving unit 162, configured to receive at least three user operations, where the user operation is an operation triggered by a user selecting a point from the image to be divided;
[0076] The static area configuration unit 163 is used to determine the area boundary points according to each user operation, and connect the area boundary points to form a polygonal area;
[0077] The dynamic adjustment unit 164 is used to dynamically adjust the area based on the clustering algorithm to form a working area and a client area.
[0078] In this embodiment, the image to be divided can be understood as an image with a region division requirement, and the image to be divided is an image obtained by shooting the entire business hall, and the image to be divided can be collected in advance. For example, an image in the business hall is collected in advance by at least one image acquisition device, and an image including the entire business hall is obtained by denoising, contrast enhancement, splicing, deduplication, etc., and this image is used as the image to be divided. The region boundary point can be understood as a point on the boundary of the region.
[0079] The image display unit 161 displays the image to be divided; the user selects a point from the image to be divided to initiate a user operation, for example, the user clicks a point in the image to be divided to trigger the user operation, and the user can click a point on the image to define the region boundary; the user operation receiving unit 162 receives at least three user operations, that is, the user selects at least three points and triggers. The user operation receiving unit 162 can capture the user operation using the mouse event capture function of OpenCV.
[0080] The static area configuration unit 163 determines the point selected by the user according to the user operation, takes this point as the area boundary point, and connects the area boundary points in sequence to form a polygonal area. Specifically, using the mouse event in OpenCV (such as cv2.setMouseCallback function), the user can click the mouse in the image window to select multiple points, and these points will be connected to form a polygon, which represents the pre-divided working area or customer area. In this way, the user can intuitively define and adjust the area boundary in the monitoring image to ensure the accuracy and practicality of the area division.
[0081] The dynamic adjustment unit 164 is used to dynamically adjust the regional division through the K-means clustering algorithm to adapt to the layout of different business halls; the K-means clustering algorithm needs to divide the data points into k clusters during calculation, and dynamically adjust the regional division by minimizing the distance from the data points in the cluster to the cluster center. The calculation formula is:
[0082]
[0083] Among them, C i is the i-th cluster, μ i is the center of the ith cluster.
[0084] The dynamic adjustment unit 164 uses this algorithm to dynamically adjust the regional division according to the actual monitoring data, so that the system can adapt to the layout changes of different business halls and improve the flexibility and accuracy of regional division. Overall, the regional division module 16 ensures the rationality and accuracy of the division of work areas and customer areas by combining static configuration and dynamic adjustment, and provides reliable basic data support for subsequent regional judgment and alarm.
[0085] Optionally, the region judgment module 14 includes: a region judgment unit 141 and a comprehensive judgment unit 142;
[0086] The area determination unit 141 is used to determine the position of the detection frame of each target object by using a ray method to determine whether the target object is in the working area. If so, the type of the target object is determined to be a staff member; otherwise, the type of the target object is determined to be a non-staff member;
[0087] The comprehensive judgment unit 142 is used to count the types of target objects in the overall image of each business hall and determine the types of target objects.
[0088] The area judgment unit 141 uses the ray method to judge the coordinates of the detection frame returned by the model to determine whether it is within the working area. Specifically, the ray method is a classic algorithm for judging whether a point is within a polygon. A ray is drawn from the point to be judged in any direction, and the number of intersections between the ray and the polygon boundary is calculated. If the number of intersections is an odd number, the point is within the polygon, otherwise it is outside the polygon. The algorithm uses simple geometric characteristics to efficiently determine whether the coordinates of the detection frame detected by the model fall within the pre-divided working area, thereby distinguishing between staff and non-staff. When the position of the detection frame is within the working area, it is determined that the target object is within the working area and the type of the target object is a staff member. Otherwise, the type of the target object is determined to be a non-staff member.
[0089] The comprehensive judgment unit 142 counts the types of target objects in each overall image of the business hall, determines the type of the same target object identified in each overall image of the business hall, and comprehensively determines the type of the target object. The comprehensive judgment unit 142 can use a time window comprehensive algorithm to comprehensively judge the detection results (i.e., the type of the target object) of multiple overall images of the business hall processed and stored in the database to improve accuracy.
[0090] Before the calculation of the time window synthesis algorithm used, the definition is first set: a sequence D containing multiple frame image detection results = {d1, d2, ..., d n}, where d i Represents the detection result of the i-th frame image, and then adopts the majority voting method, the expression is:
[0091] Final Result t =mode(d t-W+1 ,d t-W+2 ,…,d t )
[0092] Among them, Final Result t is the comprehensive judgment result at the tth frame, d t-W+1 ,d t-W+2 ,…,d t It is from The detection result sequence from frame to frame t, mode is the statistical mode function, which indicates the detection result with the most occurrences in the window.
[0093] The time window comprehensive algorithm counts the detection results of multiple frames of images within a certain time window and adopts the majority voting method (i.e., selecting the detection result that appears the most times in the window) to improve the stability and accuracy of the judgment.
[0094] Optionally, the alarm module 15 includes: a personnel information determination unit 151 and an alarm judgment unit 152;
[0095] A personnel information determination unit 151 is used to determine personnel information according to the type of the target object, where the personnel information includes at least one of the following: the number of staff members, the number of non-staff members, the working hours of staff members, and the waiting time of non-staff members;
[0096] The alarm determination unit 152 is used to determine whether the personnel information meets the corresponding alarm conditions, wherein the alarm conditions are automatically adjusted and updated according to historical data.
[0097] In this embodiment, the personnel information can be understood as information used to describe the staff and / or customers, such as the number of staff, working hours, etc.
[0098] The personnel information determination unit 151 performs statistical analysis based on the type of each target object in the overall image of each business hall to determine the personnel information. Different personnel information corresponds to different alarm conditions, and different types of personnel information have multiple corresponding alarm conditions pre-set; for example, the personnel information is the number of staff members, and the alarm condition is that the number of staff members is lower than the first preset value for a continuous period of time, and the first preset value is 1. The personnel information is the number of staff members and the number of non-staff members, and the alarm condition is that the ratio of the number of staff members to the number of non-staff members is lower than the second preset value; the personnel information is the working hours of the staff members, and the alarm condition is that the working hours of the staff members are less than the working hours threshold; the personnel information is the waiting time of non-staff members, and the alarm condition is that the waiting time of non-staff members exceeds the customer waiting time threshold; the personnel information is the number of non-staff members, and the alarm condition is that the number of non-staff members exceeds the customer number threshold.
[0099] The alarm determination unit 152 determines whether the corresponding alarm condition is met based on the personnel information. When the personnel information is multiple types of information, it can determine whether to trigger an alarm based on the judgment results of multiple dimensions. For example, when only one of the above conditions is met, an alarm is triggered, or when half of the above conditions are met, an alarm is triggered.
[0100] Among them, the alarm conditions can be automatically adjusted and updated based on historical data. Take the alarm condition that no staff can be detected for a period of time as an example, that is, the time threshold for the initial failure to detect staff can be set; the time threshold for the initial failure to detect staff can be a fixed value set when the system starts, which is used to determine whether an alarm needs to be triggered if no staff is detected for a continuous period of time. For example, an initial threshold can be set to 5 minutes. If the system does not detect a staff member within 5 minutes, an alarm is triggered. The setting of this threshold is based on experience and business needs to ensure that potential staff absences are identified and responded to in the shortest time.
[0101] By statistically analyzing historical data and dynamically adjusting the initial threshold, the alarm rules can be more adapted to the actual situation. Specific methods can include calculating the mean and standard deviation of historical data, and automatically adjusting the time threshold based on these statistics, such as setting the new threshold to the historical mean plus two times the standard deviation. This dynamic adjustment mechanism can improve the flexibility and adaptability of the system, reduce false positives and false negatives, and ensure the rationality and effectiveness of the alarm rules.
[0102] By combining data from multiple dimensions, the operation of the business hall can be evaluated more comprehensively and accurately, thereby improving the accuracy and rationality of alarm decisions. Overall, the alarm module 15 ensures that the system can accurately and timely identify and respond to various abnormal situations through the combination of rule setting, adaptive adjustment and multi-dimensional rules, and provides reliable alarm information for managers.
[0103] Optionally, the alarm module 15 includes: a first notification unit 153, a second notification unit 154 and a third notification unit 155;
[0104] A first notification unit 153 is used to automatically log in to at least one related system through an automated script and send a first alarm message after the alarm condition is met;
[0105] The second notification unit 154 is used to send the second alarm information through at least one notification channel, where the notification channel is: SMS, phone or internal communication tool;
[0106] The third notification unit 155 is used to monitor the sending status of the first alarm information and the second alarm information, and switch to the backup channel to resend if at least one of the first alarm information and the second alarm information fails to be sent.
[0107] In this embodiment, the automation script may be a Robotic Process Automation script, ie, an RPA script. The first alarm information and the second alarm information are both information for generating an alarm.
[0108] The first notification unit 153 simulates manual operation through RPA technology, and automatically logs in to the relevant system and sends notifications through RPA scripts after the alarm conditions are met; RPA is a software technology that can simulate manual operation to perform a series of predetermined tasks. In this system, when it is detected that the alarm conditions are met, the RPA notification unit triggers an RPA script, which automatically logs in to a preset management system or platform, such as a mail system, an internal communication system, etc., and then automatically fills in relevant fields and sends notifications according to the content of the alarm information. In this way, the RPA notification unit can greatly reduce the time and errors of manual operation and improve the efficiency and accuracy of notifications.
[0109] The second notification unit 154 sends the second alarm information through at least one pre-set notification channel to make an alarm notification. Among them, SMS is a direct and reliable notification method, which can ensure that the notification can still be received without a network connection. The internal communication tool can be eLink, etc. eLink is an instant communication tool within the enterprise, which is suitable for quickly spreading important information within the enterprise. The second notification unit 154 will select a suitable notification channel according to the importance and urgency of the alarm information, and send the alarm information through a preset API interface or service platform to ensure that the notification can cover all relevant personnel.
[0110] The third notification unit 155 implements a notification backup function, monitors the sending status of the first alarm information and the second alarm information, and automatically switches to the backup channel to resend if the above alarm information fails to be sent.
[0111] In actual applications, some notification channels may be temporarily unavailable or fail. In order to ensure that the alarm information can be delivered in time, the third notification unit 155 monitors the status of each notification sending. If it is detected that the main notification channel fails to send, the notification backup unit will immediately start the backup channel to resend the notification. For example, if the notification fails to be sent via SMS, the system will automatically switch to eLink or other backup channels to continue sending until it is confirmed that the notification is successfully delivered. This mechanism not only improves the reliability of notifications, but also ensures that information can be conveyed in a timely manner in an emergency, and will not be delayed due to the failure of a single channel.
[0112] The alarm module 15 combines RPA technology, multi-channel mechanism and notification backup mechanism to ensure that alarm information can be quickly and accurately conveyed to relevant managers, providing important guarantees for the stable operation and timely response of the system.
[0113] Optionally, the monitoring module 11 and the image preprocessing module 12 are connected to a network, the image preprocessing module 12 and the detection module 13 are connected to a network, the detection module 13 and the region judgment module 14 are connected to a network, the region division module 16 and the region judgment module 14 are electrically connected, and the alarm module 15 and the region judgment module 14 are connected to a network.
[0114] The business hall dynamic monitoring system provided by this application can be implemented based on big data. The business hall dynamic monitoring system provided by this application has the following beneficial effects:
[0115] 1. This application can automatically collect and flexibly determine whether to trigger an alarm based on preset and adaptively adjusted multi-dimensional rules through the cooperation of the monitoring module, image preprocessing module, detection module, area judgment module, and alarm module, ensuring that the system can provide timely warnings and personnel deployment when abnormalities occur.
[0116] 2. This application adopts the advanced YOLO model for human body detection. The pre-trained model is loaded through the model loading unit. The detection unit uses the convolutional neural network in the YOLO model to process the image and accurately identify people and objects. The hardware acceleration unit accelerates the processing through the GPU, which significantly improves the detection speed and real-time performance. It is suitable for high-concurrency scenarios. Through these technical means, the present invention can accurately identify and detect people in complex environments, and improve the accuracy and efficiency of detection.
[0117] 3. The area division module of this application can adapt to the layout changes of different business halls by combining static area configuration and dynamic adjustment. The static area configuration unit allows users to intuitively define area boundaries on the image through the mouse event capture function of OpenCV. The dynamic adjustment unit uses the K-means clustering algorithm to dynamically adjust the area division according to real-time data, ensuring the flexibility and accuracy of the area division, and providing a reliable basis for subsequent area judgment.
[0118] 4. The area judgment module of the present application improves the accuracy of judgment by combining the area judgment unit and the comprehensive judgment unit. The area judgment unit uses the ray method to accurately judge whether the human body is in the specified area. The comprehensive judgment unit uses the time window comprehensive algorithm to perform statistics and majority voting on the detection results of multiple frames of images, thereby improving the stability and accuracy of the judgment results and reducing the possibility of single-frame misjudgment.
[0119] 5. The alarm module of this application combines rule setting, adaptive adjustment and multi-dimensional rules to ensure the flexibility and rationality of the alarm rules. The initial rule setting provides a basic alarm mechanism. The adaptive adjustment unit dynamically optimizes the rules according to historical data. The multi-dimensional rule unit considers multiple factors (such as working hours, customer waiting time, number of customers, etc.) to comprehensively judge whether to trigger an alarm. Through such a design, false alarms and missed alarms can be effectively reduced, ensuring the accuracy and timeliness of alarms.
[0120] 6. The alarm module of this application combines RPA technology, multi-channel mechanism and notification backup mechanism to ensure that the alarm information can be quickly and accurately conveyed to relevant managers. The RPA notification unit simulates manual operation and automatically sends notifications, reducing the time and errors of manual operation. The multi-channel notification unit sends alarm information through multiple channels such as SMS and eLink to ensure the coverage of notifications. The notification backup unit monitors the notification sending status and automatically switches to the backup channel when the main channel fails to send, ensuring that the notification is delivered in time and improving the reliability of the system.
[0121] Embodiment 3
[0122] Figure 3This is a flow chart of a method for dynamic monitoring of a business hall provided by Embodiment 3 of the present invention. Figure 3 As shown, the method is executed by the business hall dynamic monitoring system and includes the following steps:
[0123] S201, obtaining a video stream collected by at least one image acquisition device, and performing image capture on each video stream to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each video stream, and each image acquisition device has different posture information, and the posture information includes at least one of a position and an angle;
[0124] S202, stitching each frame of the image to be processed to obtain an overall image of the business hall;
[0125] S203, detecting the entire image of the business hall to obtain a detection frame of the target object;
[0126] S204, determining the type of each target object according to the pre-divided working area and the detection frame of the target object, where the types of the target objects at least include staff and non-staff;
[0127] S205: Determine whether a preset alarm condition is met according to the type of each target object. If so, issue an alarm according to the preset alarm rule to deploy staff.
[0128] An embodiment of the present invention provides a method for dynamic monitoring of a business hall, which obtains an image data set by intercepting a video stream collected by an image acquisition device with different posture information, obtains an overall image of the business hall by splicing the images to be processed in the image data set, and realizes all-round monitoring of the business hall. By detecting the overall image, a detection frame of the target object is obtained, and the type of the target object is judged according to a pre-divided work area to determine whether the target object is a staff member or a non-staff member. When the alarm condition is met, an alarm is issued in time according to the alarm rules, and personnel are deployed to avoid the situation where customers wait too long when handling business and staff are away from their posts for a long time, thereby improving customer experience, improving customer service efficiency and satisfaction, and improving the operation management and service quality of the business hall.
[0129] Optionally, obtaining a video stream captured by at least one image capture device and performing image capture on each of the video streams includes:
[0130] Acquire the video stream through the uniform resource locator of each image acquisition device using the real-time streaming protocol;
[0131] When the image capture condition is met, a frame of the image to be processed at the same time in each of the video streams is captured.
[0132] Optionally, the images to be processed in each of the image data sets are stitched together to obtain an overall image of the business hall, including:
[0133] De-noising each of the to-be-processed images by Gaussian filtering, and contrast-enhancing each of the de-noised images by histogram equalization to obtain an enhanced image;
[0134] The enhanced images are subjected to boundary stitching and deduplication processing according to the corresponding position information to obtain an overall image of the business hall.
[0135] Optionally, detecting the overall image of each business hall to obtain a detection frame of the target object includes:
[0136] Loading a model weight file and a configuration file of a pre-trained object detection model, and completing the construction of the object detection model based on the model weight file and the configuration file;
[0137] The overall image of the business hall is processed by the object detection model, the overall image of the business hall is divided into at least one grid, the detection frame and class probability predicted for each grid are determined, and the detection frame of the target object is determined based on the detection frame and class probability.
[0138] Optionally, the method further includes: receiving a user operation, and performing area division based on the user operation to form a work area and a client area.
[0139] Optionally, receiving a user operation, and performing area division based on the user operation to form a work area and a client area, includes:
[0140] Displaying the image to be divided, wherein the image to be divided is an image obtained by photographing the entire business hall;
[0141] receiving at least three user operations, where the user operations are operations triggered by a user selecting a point from the image to be divided;
[0142] Determine area boundary points according to each of the user operations, and connect the area boundary points to form a polygonal area;
[0143] The areas are dynamically adjusted based on clustering algorithms to form work areas and customer areas.
[0144] Optionally, judging the type of each target object according to the pre-divided working area and the detection frame of the target object includes:
[0145] The position of the detection frame of each target object is judged by the ray method to determine whether the target object is in the working area, and if so, the type of the target object is determined to be a staff member, otherwise, the type of the target object is determined to be a non-staff member;
[0146] The types of the target objects in the overall images of the business halls are counted to determine the types of the target objects.
[0147] Optionally, judging whether a preset alarm condition is met according to the type of each target object includes:
[0148] Determine personnel information according to the type of the target object, the personnel information including at least one of the following: the number of staff members, the number of non-staff members, the working hours of staff members, and the waiting time of non-staff members;
[0149] Determine whether the personnel information meets corresponding alarm conditions, wherein the alarm conditions are automatically adjusted and updated according to historical data.
[0150] Optionally, alarms are issued based on pre-set alarm rules, including:
[0151] After the alarm condition is met, automatically log in to at least one related system through an automated script and send a first alarm message;
[0152] Sending the second alarm information via at least one notification channel, wherein the notification channel is: SMS, phone call or internal communication tool;
[0153] Monitor the sending status of the first alarm information and the second alarm information, and if at least one of the first alarm information and the second alarm information fails to be sent, switch to the backup channel for resending.
[0154] The business hall dynamic monitoring method provided in the embodiment of the present invention can be executed by the business hall dynamic monitoring system provided in any embodiment of the present invention, and has corresponding beneficial effects.
[0155] Embodiment 4
[0156] Figure 4A schematic diagram of the structure of an electronic device 40 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.
[0157] like Figure 4 As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.
[0158] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0159] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as a business hall dynamic monitoring method.
[0160] In some embodiments, the business hall dynamic monitoring method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the business hall dynamic monitoring method described above can be performed. Alternatively, in other embodiments, the processor 41 can be configured to execute the business hall dynamic monitoring method in any other appropriate manner (for example, by means of firmware).
[0161] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0162] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0163] An embodiment of the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method for dynamically monitoring a business hall described in any embodiment of the present invention is implemented.
[0164] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0165] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).
[0166] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.
[0167] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.
[0168] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.
[0169] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A dynamic monitoring system for business halls, It is characterized in that include: Monitoring module, image preprocessing module, detection module, area judgment module and alarm module; The monitoring module is used to obtain a video stream collected by at least one image acquisition device, and to capture images of each of the video streams to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each of the video streams, and each of the image acquisition devices has different posture information, and the posture information includes at least one of a position and an angle; The image preprocessing module is used to splice the images to be processed in each of the image data sets to obtain an overall image of the business hall; The detection module is used to detect the overall image of each business hall and obtain a detection frame of the target object; The area determination module is used to determine the type of each target object according to the pre-divided working area and the detection frame of the target object, and the types of the target objects at least include staff and non-staff; The alarm module is used to determine whether a preset alarm condition is met according to the type of each target object. If so, an alarm is issued according to the preset alarm rules to deploy staff.
2. The system according to claim 1, It is characterized in that The monitoring module includes: a video stream acquisition unit and an image capture unit; A video stream acquisition unit, used for acquiring a video stream through the uniform resource locator of each of the image acquisition devices using a real-time streaming protocol; The image capture unit is used to capture a frame of the to-be-processed image at the same time in each of the video streams when the image capture condition is met.
3. The system according to claim 1, It is characterized in that The image preprocessing module includes an image processing unit and an image stitching unit; The image processing unit is used to perform denoising on each of the to-be-processed images by Gaussian filtering, and to perform contrast enhancement on each of the denoised images by histogram equalization to obtain an enhanced image; The image stitching unit is used to perform boundary stitching and deduplication processing on each of the enhanced images according to the corresponding posture information to obtain an overall image of the business hall.
4. The system according to claim 1, It is characterized in that The detection module includes: a model loading unit and a detection unit; The model loading unit is used to load a model weight file and a configuration file of a pre-trained object detection model, and complete the construction of the object detection model based on the model weight file and the configuration file; The detection unit is used to process the overall image of the business hall through the object detection model, divide the overall image of the business hall into at least one grid, determine the detection frame and class probability predicted for each grid, and determine the detection frame of the target object based on the detection frame and class probability.
5. The system according to claim 1, It is characterized in that Also includes: Region division module; The area division module is used to receive user operations and perform area division based on the user operations to form a working area and a client area.
6. The system according to claim 5, It is characterized in that The area division module includes: an image display unit, a user operation receiving unit, a static area configuration unit and a dynamic adjustment unit; The image display unit is used to display the image to be divided, and the image to be divided is an image obtained by shooting the entire business hall; The user operation receiving unit is used to receive at least three user operations, where the user operation is an operation triggered by a user selecting a point from the image to be divided; The static area configuration unit is used to determine area boundary points according to each of the user operations, and connect the area boundary points to form a polygonal area; The dynamic adjustment unit is used to dynamically adjust the area based on a clustering algorithm to form a working area and a client area.
7. The system according to claim 1, It is characterized in that The region judgment module includes: a region judgment unit and a comprehensive judgment unit; The area judgment unit is used to judge the position of the detection frame of each target object by ray method to determine whether the target object is in the working area, and if so, determine the type of the target object is a staff member, otherwise, determine the type of the target object is a non-staff member; The comprehensive judgment unit is used to count the types of target objects in the overall image of each business hall and determine the types of the target objects.
8. The system according to claim 1, It is characterized in that The alarm module includes: a personnel information determination unit and an alarm judgment unit; A personnel information determination unit, configured to determine personnel information according to the type of the target object, wherein the personnel information includes at least one of the following: the number of staff members, the number of non-staff members, the working hours of staff members, and the waiting time of non-staff members; The alarm judgment unit is used to judge whether the personnel information meets the corresponding alarm conditions, wherein the alarm conditions are automatically adjusted and updated according to historical data.
9. The system according to claim 1, It is characterized in that The alarm module includes: a first notification unit, a second notification unit and a third notification unit; A first notification unit, configured to automatically log into at least one related system through an automated script and send a first alarm message after an alarm condition is met; The second notification unit is used to send the second alarm information through at least one notification channel, and the notification channel is: SMS, phone or internal communication tool; The third notification unit is used to monitor the sending status of the first alarm information and the second alarm information, and if at least one of the first alarm information and the second alarm information fails to be sent, switch to the backup channel for resending.
10. A method for dynamic monitoring of a business hall, It is characterized in that The method is performed by the business hall dynamic monitoring system according to any one of claims 1 to 9, including: Acquire a video stream acquired by at least one image acquisition device, and perform image capture on each of the video streams to obtain an image data set at at least one moment, wherein the image data set includes an image to be processed captured from each of the video streams, and each of the image acquisition devices has different posture information, and the posture information includes at least one of a position and an angle; Splicing each frame of the image to be processed to obtain an overall image of the business hall; Detecting the entire image of the business hall to obtain a detection frame of the target object; Determine the type of each target object according to the pre-divided working area and the detection frame of the target object, wherein the type of the target object at least includes staff and non-staff; It is determined whether the preset alarm conditions are met according to the type of each target object. If so, an alarm is issued according to the preset alarm rules to deploy staff.
