A reporting system for the absence of a product bin in leased equipment

Abstract

The present invention relates to the technical field of reporting systems, and specifically to a reporting system for the lack of product warehouses in leased equipment, which includes a data acquisition module, a data calculation module, and an information reporting module. In the data calculation module, the leased equipment is divided into regions, and there are m leased equipment in each region; for the reporting system for the lack of product warehouses in the leased equipment of the present invention, the number of missing product warehouses in the leased equipment is combined with the population density in the region where the leased equipment is located for weight measurement to obtain a comparison value, and reporting is carried out through the comparison value, which can ensure that regions with a larger number of missing items and a larger population density have a higher priority, while regions with a larger number of missing items and a smaller population density have a lower priority, ensuring more efficient operation of the leased equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of reporting systems, and more particularly to a product bin missing reporting system in rental equipment. Background Art

[0002] Rental equipment, the most common in the current market are various public rental and sharing equipment such as shared power banks and shared bicycles. Taking shared power banks as an example, when a power bank is rented and used, the product bin shows a shortage. Facing a large amount of shortage data, in the current technology, the background will divide by region, sort out the shortage situations in each region where the power bank rental equipment is located, and dispatch maintenance personnel to replenish the region with the largest number of shortages; this method has problems of low efficiency in actual use. Because although some regions have a large number of shortages, the current population density is not high, and the urgency for replenishment is not high. While in some regions, the number of shortages is slightly less than that of more regions, but the population density in the region is greater. The greater the population density, the higher the possibility of passing by the rental equipment and the higher the possibility of using the rental equipment. Therefore, the urgency for replenishment is higher. The existing reporting system clearly cannot report more efficiently. Summary of the Invention

[0003] The purpose of the present invention is to provide a product bin missing reporting system in rental equipment to solve the problems raised in the above background art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: A product bin missing reporting system in rental equipment, including a data acquisition module, a data calculation module, and an information reporting module. In the data calculation module, the rental equipment is divided into regions, with m rental equipment in each region. Let the number of missing product bins in each rental equipment be , where , and let the total shortage in each region be , then , where t is an integer greater than or equal to 1, representing different region numbers;

[0005] The data acquisition module includes obtaining the real-time numerical value of the number of missing product bins in each rental equipment , and obtaining the real-time numerical value of the population flow in each of the above regions. Let the numerical value of the population flow obtained in a certain region be , and let the corresponding region area be , is known, calculate the regional population density. Let the population density be , then , that is is the ratio of , and As a weight coefficient, calculate the comparison value , , where t is an integer greater than or equal to 1, representing different area numbers. When the values of t are the same, it means they correspond to the same area;

[0006] Arrange the comparison values in descending order. According to the arrangement of , obtain the corresponding area arrangement order. Let the simultaneous area maintenance quantity be G. Report the areas ranked in the top G in the area arrangement order through the information reporting module, and at the same time report the total quantity missing in the corresponding areas , and dispatch maintenance personnel to make up for it. The value of G is set according to the number of maintenance personnel.

[0007] Preferably, the data acquisition module obtains the numerical value of the pedestrian flow in the area through a camera. The specific steps are as follows: Connect the cameras to the background server uniformly through network cables. The server periodically sends time synchronization signals to all cameras in the area. After receiving the signals, the cameras adjust their internal clocks to synchronize the image frames captured at the same moment within the error range; The cameras continuously capture RGB format video streams, divide them into independent image frames according to the established frame rate, and transmit each frame to the server for processing.

[0008] Preferably, a color image contains three RGB channels. Map the three-dimensional color space to a one-dimensional gray space through gray conversion to reduce the amount of calculation; Then perform noise reduction filtering and edge enhancement processing in sequence, and then perform human body contour detection.

[0009] Preferably, in the initialization stage of the human body contour detection, collect multiple frames of background images without humans, establish K Gaussian distribution models for each pixel point, with distribution having mean, covariance, and weight parameters; When a new frame arrives, compare the pixel value with the model. If it matches, update the corresponding Gaussian distribution parameters. If it doesn't match, it is a foreground target. Sort the Gaussian distributions by weight. If the sum of weights is greater than the set value, it is the background, and the rest are the foreground. Dynamically update the model to adapt to the gradual change of light.

[0010] Preferably, for contour extraction, for the foreground image, the findContours function in OpenCV is based on the contour topological traversal algorithm. Starting from the image boundary, trace the contour boundary pixels according to connectivity and mark all closed contours; Set screening criteria. The area threshold is set according to the estimated minimum and maximum sizes of the human body in the scene, and the perimeter threshold is used to assist in excluding irregular redundant contours; Consider the contour compactness through the shape factor, and exclude those that do not conform to the human body aspect ratio and ellipticity to initially screen out suspected human body contours.

[0011] Preferably, calculate the aspect ratio of the suspected human contour. The long axis and short axis are determined according to the circumscribed rectangle of the contour. The circularity measures the degree of approximation of the contour to a circle and is evaluated by the ratio of the square of the contour perimeter to the area. Obtain the data for labeled human and non-human contours through the above feature evaluation. Train a support vector machine with the labeled human and non-human contour data. Input the extracted feature vectors into the trained model and output the discrimination result. The positive class is the human contour, and an accurate human contour set is constructed.

[0012] Preferably, place a standard checkerboard calibration object on the regional ground. The camera takes pictures from different perspectives. Use OpenCV functions to detect the checkerboard corner points. Based on the corner point coordinates, calculate the internal and external parameters of the camera through the Zhang Zhengyou calibration algorithm. After obtaining the internal and external parameters, transform the human contour coordinates in different camera coordinate systems to a unified world coordinate system through the coordinate transformation formula.

[0013] Preferably, construct a cost matrix. The rows represent the human contours of one camera, the columns are the contours of another camera, and the elements are the quantification values of the differences in their positions, sizes, and movement directions. Find the minimum-cost matching through the Hungarian algorithm to associate the same human contour captured by different cameras, eliminate duplicate counting, and summarize the human contours after deduplication. The quantity is the estimated value of the current number of people in the region.

[0014] Compared with the prior art, the beneficial effects of the present invention are:

[0015] In the product bin missing report system of the rental equipment of the present invention, the missing quantity of the product bin in the rental equipment is combined with the population density of the area where the rental equipment is located for weight measurement to obtain a comparison value, and reports are made through the comparison value. It can ensure that the areas with a larger missing quantity and a larger population density have a higher priority, while the areas with a larger missing quantity and a smaller population density have a lower priority, ensuring the more efficient operation of the rental equipment.

[0016] In the present invention, the method for the data acquisition module to obtain the numerical value of the number of people in the region through the camera can more efficiently and accurately estimate the number of people in the region, is applicable to open areas where it is difficult to count the number of people entering and leaving, and realizes the statistics of the number of people in the region. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the product bin missing report system of the rental equipment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] Please refer to Figure 1 , the present invention provides a technical solution: a product bin missing reporting system in a rental device, including a data acquisition module, a data calculation module, and an information reporting module. The data calculation module includes a central processing unit and a readable storage medium storing a computer program. The computer program is executed through the central processing unit. In the data calculation module, the rental devices are divided into regions. Taking shared power banks as an example, there are several shared power bank rental devices located within the map range. The map is divided into regions to ensure that the area and the number of rental devices in each region do not vary too much. There are m rental devices in each region. Let the number of missing product bins in each rental device be , where , let the total amount missing in each region be , then , where t is an integer greater than or equal to 1, representing different region numbers;

[0020] The data acquisition module includes obtaining the real-time numerical value of the number of missing product bins in each rental device , and obtaining the real-time numerical value of the number of people in each of the above regions. The data acquisition module includes sensors arranged at the corresponding positions of the product bins in the rental devices. Through the sensors in the product bins, it can be detected in real time whether the products in the product bins are missing. The above sensors can adopt contact switch types, photoelectric switch types, and capacitive sensors, etc. When the products in the product bins are taken away, a signal can be fed back to detect the missing.

[0021] Let the numerical value of the number of people obtained in a certain region be , let the corresponding region area be , is known. Since the regions are divided on the map, the region area can be calculated according to the map ratio. Calculate the regional population density. Let the population density be , then , that is is and ratio, take as the weight coefficient, calculate the comparison value , , where t are all integers greater than or equal to 1, representing different region numbers. When the values of t are the same, it means corresponding to the same region;

[0022] Arrange the comparison values in descending order. According to this arrangement, obtain the corresponding area arrangement order. Assume the number of areas maintained simultaneously is G. Report the top G areas in the area arrangement order through the information reporting module, and at the same time report the total quantity missing in the corresponding areas , and dispatch maintenance personnel to make up for it. The value of G is set according to the number of maintenance personnel. For example, if the maintenance personnel can only be divided into 3 groups for maintenance and replenishment at the same time, then the value of G is set to 3 in the system. When reporting, only the top three areas will be reported, and 3 groups of maintenance personnel will be dispatched to the corresponding areas for maintenance and replenishment; the information reporting module consists of an Internet of Things communication system and mainly plays a communication role.

[0023] The data acquisition module obtains the numerical value of the number of people in the area. When the area is an enclosed place, such as a shopping mall, a device for counting the number of people can be set at the entrance and exit for statistics of the incoming and outgoing people; while in places such as open streets, it is impossible to achieve incoming and outgoing statistics, and several cameras need to be set in the area, trying to avoid dead angles. The cameras are uniformly connected to the background server through network cables. The server periodically sends time synchronization signals to all cameras in the area. After receiving the signals, the cameras adjust their internal clocks so that the image frames captured at the same moment by different cameras are synchronized within the error range; the cameras continuously capture RGB format video streams, divide them into independent image frames at a predetermined frame rate, and transmit each frame to the server for processing. The transmission protocol can be selected from HTTP, RTSP, etc.

[0024] The color image contains three RGB channels, and the data redundancy of the three channels is high. Map the three-dimensional color space to a one-dimensional gray space through gray conversion to reduce the amount of calculation; then perform noise reduction filtering and edge enhancement processing in sequence, and then perform human contour detection. Noise reduction filtering and edge enhancement processing are both common image processing techniques. Gaussian filtering can be used for noise reduction filtering to generate a convolution kernel based on the Gaussian function, and use this convolution kernel to perform convolution operations on each pixel of the image. The central pixel value is updated to the weighted sum of the neighborhood pixels, and the abrupt pixel values caused by noise are smoothed, improving the image quality and facilitating subsequent accurate edge extraction. For edge enhancement, first smooth the image with Gaussian filtering, and then calculate the gradient magnitude and direction using Sobel-like operators. After non-maximum suppression, only the strongest edge points in the gradient direction are retained; then through double-threshold detection, those higher than the high threshold are retained, those lower than the low threshold are discarded, and the intermediate values are retained depending on the neighborhood situation. Finally, connect the edge breakpoints to obtain a coherent and clear human contour.

[0025] In the initialization stage of human body contour detection, multiple background images without humans are collected, and K Gaussian distribution models are established for each pixel. The Gaussian distribution model is a widely used probability distribution model with mean, covariance, and weight parameters. When a new frame arrives, the pixel value is compared with the model. If it matches, the corresponding Gaussian distribution parameters are updated; if not, it is a foreground object. The Gaussian distributions are sorted by weight, and if the sum of weights is greater than a set value, it is considered background, otherwise it is foreground. The model is dynamically updated to adapt to gradual changes in light.

[0026] For contour extraction, for the foreground image, the findContours function in OpenCV is based on the contour topological traversal algorithm. Starting from the image boundary, it traces the contour boundary pixels according to connectivity and marks all closed contours. findContours is a key function in the OpenCV library for detecting object contours in images. It can accurately locate the boundaries of target objects in a properly preprocessed binary image. The contour topological traversal algorithm on which the findContours function in OpenCV is based focuses on tracing the contour boundaries of target objects in the image. By systematically exploring the connection relationships of pixel points, it outlines closed or non-closed contours. The function starts searching from the boundary pixels of the image. In a binary image obtained after image preprocessing (where the target is white and the background is black, or vice versa), it moves along the boundary where the pixel values change. For example, it starts exploring from the junction of a white foreground object and a black background, and the pixel at this junction is the starting tracking point. Using the 8-connectivity (in a two-dimensional plane, a pixel has 8 adjacent pixels, including diagonal directions) or 4-connectivity (only considering horizontally and vertically adjacent pixels) rules for pixels, it determines which pixels belong to the same connected region, that is, the same contour.

[0027] If the pixel values of adjacent pixels satisfy the conversion relationship between foreground and background and meet the connectivity setting, they are considered to belong to the same contour. Once the starting point is found, the algorithm uses a recursive or iterative method to continuously expand the tracking path to surrounding pixels. Taking the recursive method as an example, when a boundary pixel is located, the function calls itself to check the adjacent pixels of this pixel and marks the qualified adjacent pixels as part of the contour, continuously delving deeper until it returns to the starting point to form a closed contour. The iterative method uses a stack structure to push the pixels to be explored onto the stack and checks the adjacent pixels of the popped pixels one by one, continuously updating the stack content until the contour tracking is completed.

[0028] Set screening criteria. The area threshold is set according to the estimated minimum and maximum sizes of the human body in the scene, and the perimeter threshold is used to assist in excluding irregular redundant contours. The shape factor is used to consider the compactness of the contour, and those that do not meet the human body aspect ratio and ellipticity are excluded to initially screen out suspected human body contours.

[0029] Calculate the aspect ratio of the suspected human contour. The long axis and short axis are determined according to the circumscribed rectangle of the contour. The circularity measures the degree to which the contour approaches a circle and is evaluated by the ratio of the square of the contour perimeter to the area. The data of the labeled human and non-human contours are obtained through the above feature evaluation. Train a support vector machine with the labeled human and non-human contour data, input the extracted feature vectors into the trained model, and output the discrimination result. The positive class is the human contour, and an accurate human contour set is constructed.

[0030] Place a standard checkerboard calibration object on the regional ground. The camera takes pictures from different perspectives. Use OpenCV functions to detect the checkerboard corner points. Based on the corner point coordinates, calculate the internal and external parameters of the camera through the Zhang Zhengyou calibration algorithm. The Zhang Zhengyou calibration algorithm is a commonly used camera calibration method in the field and will not be elaborated here. After obtaining the internal and external parameters, transform the human contour coordinates in different camera coordinate systems to a unified world coordinate system through the coordinate transformation formula.

[0031] Construct a cost matrix. The rows represent the human contours of one camera, the columns are the contours of another camera, and the elements are the quantization values of the differences in their positions, sizes, and movement directions. Find the minimum-cost matching through the Hungarian algorithm to associate the same human contours captured by different cameras, eliminate duplicate counting, and summarize the de-duplicated human contours. The quantity is the estimated value of the current number of people in the region. The Hungarian algorithm is an existing efficient combinatorial optimization algorithm for solving the assignment problem, which is used to optimally match the elements in two sets to minimize the matching cost.

[0032] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A reporting system for the absence of a product warehouse in leased equipment, comprising a data acquisition module, a data calculation module, and an information reporting module, characterized in that: In the data calculation module, the leased equipment is divided into regions, with m leased equipment in each region. Let the number of missing product warehouses in each leased equipment be , where . Let the total amount of missing in each region be , then , where t is an integer greater than or equal to 1, representing different region numbers; The data acquisition module includes the quantity missing in the product warehouse of each leased device for real-time numerical acquisition, and real-time numerical acquisition of the number of people in each of the above regions. Let the numerical value of the number of people in a certain region be , and let the corresponding area of the region be , be known, calculate the regional population density. Let the population density be , then , that is is the ratio of . Take as the weight coefficient, calculate the comparison value , , where t is an integer greater than or equal to 1, representing different region numbers. When the values of t are the same, it means corresponding to the same region; Compare values Arrange them in descending order. According to the arrangement, obtain the corresponding area arrangement order. Let the simultaneous area maintenance quantity be G. Report the top G areas in the area arrangement order through the information reporting module, and at the same time report the total quantity missing in the corresponding areas , dispatch maintenance personnel for replenishment. The value of G is set according to the number of maintenance personnel.

2. The product bin missing reporting system in a rental device according to claim 1, characterized in that: The described data acquisition module obtains the numerical value of the pedestrian flow in the area through a camera. The specific steps are as follows: The cameras are uniformly connected to the background server through network cables. The server periodically sends time synchronization signals to all cameras in the area. After receiving the signals, the cameras adjust their internal clocks to synchronize the image frames captured at the same moment within the error range. The cameras continuously capture RGB-format video streams, divide them into independent image frames at a predetermined frame rate, and transmit each frame to the server for processing.

3. The product bin missing reporting system in a rental device according to claim 2, wherein: The color image contains three RGB channels. By performing gray-scale conversion, the three-dimensional color space is mapped to a one-dimensional gray-scale space to reduce the amount of calculation. Then, noise reduction filtering and edge enhancement processing are performed in sequence, and then human body contour detection is carried out.

4. The product bin missing reporting system in a rental device according to claim 3, characterized in that: In the initialization stage of the described human body contour detection, multiple background images without humans are collected, and K Gaussian distribution models are established for each pixel point, with distribution having mean, covariance, and weight parameters. When a new frame arrives, the pixel value is compared with the model. If it matches, the corresponding Gaussian distribution parameters are updated. If it does not match, it is a foreground target. The Gaussian distributions are sorted by weight. If the sum of weights is greater than the set value, it is the background, and the rest are the foreground. The model is dynamically updated to adapt to the gradual change of light.

5. The product bin missing reporting system in a rental device according to claim 4, characterized in that: Contour extraction is performed. For the foreground image, the findContours function of OpenCV is based on the contour topology traversal algorithm. Starting from the image boundary, it traces the contour boundary pixels according to connectivity and marks all closed contours. Set screening criteria. The area threshold is set according to the estimated minimum and maximum sizes of the human body in the scene, and the perimeter threshold is used to assist in excluding irregular redundant contours. The compactness of the contour is considered through the shape factor, and those that do not meet the human body aspect ratio and roundness are excluded, and the suspected human body contours are initially screened.

6. The product bin missing reporting system in a rental device according to claim 5, characterized in that: Calculate the aspect ratio of the suspected human body contour. The major axis and minor axis of the aspect ratio are determined according to the bounding rectangle of the contour. The circularity measures the degree to which the contour approaches a circle and is obtained by evaluating the ratio of the square of the contour perimeter to the area. The data of labeled human and non-human contours are obtained through the above feature evaluation. The support vector machine is trained with the labeled human and non-human contour data, and the extracted feature vectors are input into the trained model to output the discrimination result. The positive class is the human body contour, and an accurate human body contour set is constructed.

7. A product bin missing reporting system in a rental device, characterized in that: Place a standard checkerboard calibration object on the ground in the area. The camera takes pictures from different perspectives, uses the OpenCV function to detect the checkerboard corner points, and calculates the internal and external parameters of the camera through the Zhang Zhengyou calibration algorithm based on the corner point coordinates. After obtaining the internal and external parameters, the human body contour coordinates in different camera coordinate systems are transformed to a unified world coordinate system through the coordinate transformation formula.

8. The product bin missing reporting system in a rental device according to claim 7, wherein: Construct a cost matrix. The rows represent the human body contours of one camera, the columns are the contours of another camera, and the elements are the quantization values of the differences in their positions, sizes, and movement directions. The Hungarian algorithm is used to find the minimum cost matching to associate the same human body contours captured by different cameras, eliminate duplicate counting, and summarize the deduplicated human body contours. The quantity is the estimated value of the current number of people in the area.

Images