Non-motor vehicle scheduling method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a non-motor vehicle scheduling method, and the method comprises the steps: obtaining a collection image in a target region, carrying out the recognition of a non-motor vehicle in the collection image based on a preset target recognition algorithm, obtaining a target detection frame corresponding to the non-motor vehicle, and carrying out the scheduling of the non-motor vehicle based on the target detection frame and the collection image. And determining a proportion and a repetition area of the target detection frame in the acquired image, determining a non-motor vehicle scheduling strategy corresponding to the target area based on the proportion and the repetition area, and performing non-motor vehicle scheduling on the target area based on the non-motor vehicle scheduling strategy. The non-motor vehicles in the collected image are accurately identified, and the proportion and the repetition area of the target detection frame in the collected image are calculated. And meanwhile, according to information such as the proportion and the repetition area, a non-motor vehicle scheduling strategy corresponding to the target area is determined, so that intelligent scheduling and intelligent management of non-motor vehicles in the target area are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of non-motor vehicle management, and in particular to a non-motor vehicle dispatching method, device, electronic equipment and storage medium. Background Art

[0002] With the acceleration of urbanization and the diversification of people's travel methods, non-motor vehicles have become one of the important tools for daily travel. Due to the large number of major non-motor vehicle manufacturers on the market, the control of non-motor vehicles is not detailed and strict enough, the deployment of non-motor vehicles is not accurate enough, and the scheduling is not smart enough, resulting in shared bicycles and motorcycles parked everywhere on the road. The existing solutions mainly use manual inspection and scheduling, which is not only inefficient, but also prone to omissions and errors. Therefore, how to provide a solution for intelligently identifying the number of non-motor vehicles and scheduling has become an urgent problem to be solved. Summary of the invention

[0003] The embodiment of the present invention provides a non-motor vehicle dispatching method, which aims to solve the problem that the existing manual inspection and dispatching method is not only inefficient but also prone to omissions and errors. The non-motor vehicles in the captured image are accurately identified, and the proportion and repeated area of ​​the target detection frame in the captured image are calculated. At the same time, according to the information such as the proportion and repeated area, the non-motor vehicle dispatching strategy corresponding to the target area is determined to realize the intelligent dispatching and intelligent management of the non-motor vehicles in the target area.

[0004] In a first aspect, an embodiment of the present invention provides a non-motor vehicle dispatching method, the method comprising:

[0005] Acquire a captured image within a target area;

[0006] Identify the non-motor vehicle in the captured image based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle;

[0007] Based on the target detection frame and the collected image, determining a proportion of the target detection frame in the collected image and a repeated area, wherein the repeated area is an area of ​​overlap between target detection frames when there are multiple target detection frames;

[0008] Based on the proportion and the repeated area, determining a non-motor vehicle dispatching strategy corresponding to the target area;

[0009] Non-motor vehicles are dispatched to the target area based on the non-motor vehicle dispatch strategy.

[0010] Optionally, before acquiring the captured image in the target area, the method further includes:

[0011] Acquire road snapshot images in the target city;

[0012] Determining the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on the captured road image;

[0013] The target area is determined based on the number of non-motor vehicles of each non-motor vehicle manufacturer and the geographical location of the target city.

[0014] Optionally, determining the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on the captured road image includes:

[0015] Identify non-motor vehicles based on the captured road image to determine the number and color of non-motor vehicles in the target city;

[0016] The number of non-motor vehicles of each non-motor vehicle manufacturer is determined based on the number and color of the non-motor vehicles in the target city.

[0017] Optionally, determining the proportion of the target detection frame in the captured image based on the target detection frame and the captured image includes:

[0018] Based on the target detection frame, determine the width and height of the target detection frame;

[0019] Based on the acquired image, determining the width and height of the acquired image;

[0020] Based on the width and height of the target detection frame and the width and height of the acquired image, the proportion of the target detection frame in the acquired image is determined.

[0021] Optionally, the target detection frame is a plurality of target detection frames, and the determining, based on the target detection frames and the acquired image, a repetitive area of ​​the target detection frames in the acquired image comprises:

[0022] Based on the target detection frames, determine the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each of the multiple target detection frames;

[0023] Based on the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each target detection frame among the multiple target detection frames, a repeated area of ​​the target detection frame in the acquired image is determined.

[0024] Optionally, the target detection frame is a plurality of target detection frames, each of the plurality of target detection frames corresponds to a proportion, and the plurality of target detection frames correspond to a plurality of proportions. Determining the non-motor vehicle dispatching strategy corresponding to the target area based on the proportions and the repeated area includes:

[0025] Determining the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas;

[0026] Comparing the number of non-motor vehicles in the target area with the preset number of non-motor vehicles in the target area to obtain a comparison result;

[0027] Based on the comparison result, a non-motor vehicle dispatching strategy corresponding to the target area is determined.

[0028] Optionally, determining the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas includes:

[0029] Based on the multiple proportions and repeated areas, determining an overlap area increment corresponding to the target area;

[0030] Based on the multiple proportions, determining an average proportion corresponding to the target area;

[0031] Based on the average proportion and the overlap area increment, the number of non-motor vehicles in the target area is determined.

[0032] In a second aspect, an embodiment of the present invention further provides a non-motor vehicle dispatching device, the non-motor vehicle dispatching device comprising:

[0033] A first acquisition module is used to acquire a captured image within a target area;

[0034] A first recognition module, configured to recognize the non-motor vehicle in the captured image based on a preset target recognition algorithm, and obtain a target detection frame corresponding to the non-motor vehicle;

[0035] A first determination module, configured to determine a proportion and a repeated area of ​​the target detection frame in the captured image based on the target detection frame and the captured image;

[0036] A second determination module is used to determine the non-motor vehicle dispatching strategy corresponding to the target area based on the proportion and the repeated area;

[0037] The first scheduling module is used to schedule non-motor vehicles in the target area based on the non-motor vehicle scheduling strategy.

[0038] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the non-motor vehicle dispatching method provided in an embodiment of the present invention when executing the computer program.

[0039] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the non-motor vehicle dispatching method provided in the embodiment of the invention are implemented.

[0040] In an embodiment of the present invention, a captured image in a target area is obtained, and non-motor vehicles in the captured image are identified based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle. Based on the target detection frame and the captured image, the proportion and repeated area of ​​the target detection frame in the captured image are determined. Based on the proportion and repeated area, a non-motor vehicle dispatching strategy corresponding to the target area is determined, and non-motor vehicles are dispatched in the target area based on the non-motor vehicle dispatching strategy. Non-motor vehicles in the captured image are accurately identified, and the proportion and repeated area of ​​the target detection frame in the captured image are calculated. At the same time, based on information such as the proportion and repeated area, the non-motor vehicle dispatching strategy corresponding to the target area is determined to achieve intelligent dispatching and intelligent management of non-motor vehicles in the target area. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0042] Figure 1 is a flow chart of a non-motor vehicle dispatching method provided by an embodiment of the present invention;

[0043] Figure 2 is a structural schematic diagram of a non-motor vehicle dispatching device provided in an embodiment of the present invention;

[0044] Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] like Figure 1 As shown, Figure 1 1 is a flow chart of a non-motor vehicle dispatching method provided by an embodiment of the present invention. The non-motor vehicle dispatching method comprises the following steps:

[0047] 101. Acquire a captured image in a target area.

[0048] In an embodiment of the present invention, the non-motor vehicle scheduling method can be deployed in a non-motor vehicle management platform. The non-motor vehicle management platform can be a server or server cluster with functions such as image processing, image recognition, data transmission, and data storage. The target area can be any area where non-motor vehicle scheduling is required. The acquired image can be a single-frame image or an image sequence composed of multiple continuous or discontinuous single-frame images. The acquired image can be stored in the non-motor vehicle management platform. The acquired image can be acquired by any electronic device that can perform image acquisition. Specifically, the acquisition device can be a drone, which takes aerial photos of the target area through the drone to obtain the acquired image; a road camera, which uses the road camera deployed in the target area to acquire the image of the target area to obtain the acquired image; a satellite, which performs satellite remote sensing photography through a satellite to obtain a satellite remote sensing image, and uses the satellite remote sensing image as the acquired image; a mobile phone or a camera, which is manually used with a mobile phone or a camera or other shooting device to capture the image to obtain the acquired image.

[0049] In a possible embodiment, when the above-mentioned drone, road camera, satellite, mobile phone or camera and other acquisition devices acquire images of the above-mentioned target area, an original image is obtained, and the above-mentioned acquisition device transmits the original image to the above-mentioned non-motor vehicle management platform based on the above-mentioned data transmission function, the above-mentioned non-motor vehicle management platform receives the above-mentioned original image, pre-processes the above-mentioned original image, obtains the above-mentioned acquired image, and stores the above-mentioned acquired image for subsequent call. The above-mentioned pre-processing can be grayscale processing, geometric transformation processing, image enhancement processing, image denoising processing and other processing methods.

[0050] 102. Identify the non-motor vehicle in the captured image based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle.

[0051] In an embodiment of the present invention, the preset target recognition algorithm may be a Faster R-CNN algorithm, which generates candidate regions through a region proposal network (RPN), and then uses a convolutional neural network (CNN) to extract the non-motor vehicle features of the above-mentioned acquisition image, and finally classifies and adjusts the position of the candidate regions through a classifier or a regressor to obtain a target detection frame corresponding to the above-mentioned non-motor vehicle, or may be a YOLO (You Only Look Once) algorithm or an SSD (Single Shot MultiBox Detector) algorithm, which performs image recognition on the above-mentioned acquisition image through the preset target recognition algorithm to obtain a target detection frame corresponding to the above-mentioned non-motor vehicle. The size of the above-mentioned target detection frame corresponds to the size of the above-mentioned non-motor vehicle. If the above-mentioned target detection frame is larger, the non-motor vehicle in the above-mentioned target detection frame is larger, and if the above-mentioned target detection frame is smaller, the non-motor vehicle in the above-mentioned target detection frame is smaller. The color corresponding to the above-mentioned target detection frame may be a fixed color, for example, black, or may also include multiple colors, and different colors correspond to different non-motor vehicle categories.

[0052] For example, if the above non-motor vehicle category is a manufacturer category, and the number of manufacturers corresponding to the above manufacturer category is three, including the first manufacturer category, the second manufacturer category, and the third manufacturer category, then the target detection frame color corresponding to the first manufacturer category is yellow, the target detection frame color corresponding to the second manufacturer category is green, and the target detection frame color corresponding to the first manufacturer category is red. If the above non-motor vehicle category is a non-motor vehicle attribute category, including non-motor vehicle attribute categories such as bicycle category, electric bicycle category, and tricycle category, then the target detection frame color corresponding to the bicycle category is blue, the target detection frame color corresponding to the electric bicycle category is orange, and the target detection frame color corresponding to the tricycle category is pink.

[0053] Specifically, a target recognition model to be trained is constructed through the above-mentioned preset target recognition algorithm, the above-mentioned target recognition model to be trained is trained to obtain a trained target recognition model, the above-mentioned collected image is input into the above-mentioned trained target recognition model, and recognition processing is performed to obtain the target detection frame corresponding to the above-mentioned non-motor vehicle.

[0054] The above-mentioned trained target recognition model can be obtained through supervised training through a large number of sample images. The above-mentioned sample images can be images containing non-motor vehicles. The labels corresponding to the images containing non-motor vehicles include the location of the non-motor vehicle and the category of the non-motor vehicle. During the training process, the sample images are input into the target recognition model to be trained, and the loss is calculated between the output results of the target recognition model to be trained and the labels corresponding to the sample images to obtain the loss function. Taking the minimum loss function as the optimization goal, the parameters of the target recognition model to be trained are adjusted through the back propagation algorithm, and the above-mentioned parameter adjustment process is iterated until the target recognition model to be trained converges at the minimum loss function, or the number of iterations reaches a preset number, the training is stopped, and a trained detection model is obtained.

[0055] In a possible embodiment, after the non-motor vehicle management platform obtains the captured image within the target area, a target recognition model to be trained is constructed based on the target recognition algorithm, the target recognition model to be trained is trained to obtain a trained target recognition model, the captured image is input into the trained target recognition model to obtain a target detection frame corresponding to the non-motor vehicle.

[0056] 103. Based on the target detection frame and the captured image, determine the proportion and repeated area of ​​the target detection frame in the captured image.

[0057] In the embodiment of the present invention, the above-mentioned proportion is the proportion of the above-mentioned target detection box in the above-mentioned acquired image, and the above-mentioned repeated area is the area of ​​the overlapping region between multiple target detection boxes when the above-mentioned target detection is multiple.

[0058] Specifically, the number of pixels in the target detection frame can be determined based on the target detection frame, the number of pixels in the captured image can be determined based on the captured image, and the proportion of the target detection frame in the captured image can be determined based on the ratio of the number of pixels in the target detection frame to the number of pixels in the captured image. Alternatively, the area of ​​the target detection frame can be determined based on the frame height and frame width of the target detection frame, the area of ​​the captured image can be determined based on the width and height of the captured image, and the proportion of the target detection frame in the captured image can be determined based on the ratio of the area of ​​the target detection frame to the area of ​​the captured image.

[0059] Specifically, if the above proportion is determined based on the above number of pixels, it can be expressed by the following formula:

[0060]

[0061] Among them, the above P iIt is expressed as the proportion of the i-th picture, the above k is expressed as the total number of target detection boxes in the i-th picture, and the above j is expressed as i It is represented as the jth target detection box in the i-th picture, Q ji It is represented as the number of pixels of the jth target detection box in the i-th image, Q i Represented as the number of pixels in the i-th image.

[0062] When the target detection frame is a plurality of target detection frames and they have overlapping parts, the overlapping area between the plurality of target detection frames may be determined according to the number of pixels between the plurality of target detection frames.

[0063] In a possible embodiment, when the non-motor vehicle management platform identifies the non-motor vehicle in the captured image based on a preset target recognition algorithm and obtains the target detection frame corresponding to the non-motor vehicle, the area of ​​the target detection frame is determined based on the height and width of the target detection frame, the area of ​​the captured image is determined based on the height and width of the captured image, the proportion of the target detection frame in the captured image is determined based on the ratio of the area of ​​the captured image to the area of ​​the target detection frame, the number of pixels of the target detection frame is determined based on the upper target detection frame, and the repeated area of ​​the target detection frame is determined based on the number of pixels in the overlapping part of the target detection frame.

[0064] 104. Based on the proportion and repeated area, determine the non-motor vehicle dispatching strategy corresponding to the target area.

[0065] In an embodiment of the present invention, the number of non-motor vehicles in the above-mentioned target area can be determined based on the above-mentioned proportion and repeated area, and the preset number of non-motor vehicles in the target area corresponding to the above-mentioned target area can be determined according to regional data such as road width, vehicle flow, pedestrian flow, geographical location, OD passenger flow data, etc. in the above-mentioned target area. Based on the above-mentioned number of non-motor vehicles and the above-mentioned preset number of non-motor vehicles in the target area, the non-motor vehicle dispatching strategy corresponding to the above-mentioned target area can be determined.

[0066] Specifically, if the number of non-motor vehicles in the target area is 200, and the preset number of non-motor vehicles in the target area corresponding to the target area is 280-300, then the non-motor vehicle dispatching strategy corresponding to the target area can be determined to dispatch at least 80 non-motor vehicles and at most 100 non-motor vehicles to the target area. If the number of non-motor vehicles in the target area is 360, and the preset number of non-motor vehicles in the target area corresponding to the target area is 280-300, then the non-motor vehicle dispatching strategy corresponding to the target area can be determined to remove at least 60 non-motor vehicles and at most 80 non-motor vehicles from the target area.

[0067] In a possible embodiment, after the non-motor vehicle management platform determines the proportion and repeated area of ​​the target detection box in the captured image, the number of non-motor vehicles in the target area is determined based on the proportion and repeated area, and based on the area data, the preset number of non-motor vehicles in the target area corresponding to the target area is analyzed, and based on the number of non-motor vehicles and the preset number of non-motor vehicles in the target area, the non-motor vehicle scheduling strategy is determined.

[0068] 105. Non-motor vehicle dispatch is performed in the target area based on the non-motor vehicle dispatch strategy.

[0069] In an embodiment of the present invention, the above-mentioned non-motor vehicle management platform can send the above-mentioned non-motor vehicle scheduling strategy to the non-motor vehicle manufacturer through the above-mentioned data transmission function. After receiving the above-mentioned non-motor vehicle scheduling strategy, the above-mentioned non-motor vehicle manufacturer matches the above-mentioned non-motor vehicle strategy according to its own non-motor vehicle reserve quantity, and dispatches the non-motor vehicle to the above-mentioned target area.

[0070] For example, when the non-motor vehicle dispatching strategy is to dispatch at least 80 non-motor vehicles and at most 100 non-motor vehicles to the target area, the non-motor vehicle reserve of the non-motor vehicle manufacturer is 90 non-motor vehicles. The non-motor vehicle reserve is matched with the non-motor vehicle dispatching strategy. If the non-motor vehicle reserve is within the scope of the non-motor vehicle dispatching strategy, the non-motor vehicle reserve can be directly dispatched to the target area.

[0071] In a possible embodiment, after the non-motor vehicle management platform determines the non-motor vehicle dispatching strategy corresponding to the target area, the non-motor vehicle dispatching strategy is sent to the non-motor vehicle manufacturer through the data transmission function. After the non-motor vehicle manufacturer receives the non-motor vehicle strategy, it matches the non-motor vehicle strategy with its own non-motor vehicle reserve. After the match is successful, the non-motor vehicle dispatch is performed on the target area.

[0072] In an embodiment of the present invention, a captured image in a target area is obtained, and non-motor vehicles in the captured image are identified based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle. Based on the target detection frame and the captured image, the proportion and repeated area of ​​the target detection frame in the captured image are determined. Based on the proportion and repeated area, a non-motor vehicle dispatching strategy corresponding to the target area is determined, and non-motor vehicles are dispatched in the target area based on the non-motor vehicle dispatching strategy. Non-motor vehicles in the captured image are accurately identified, and the proportion and repeated area of ​​the target detection frame in the captured image are calculated. At the same time, based on information such as the proportion and repeated area, the non-motor vehicle dispatching strategy corresponding to the target area is determined to achieve intelligent dispatching and intelligent management of non-motor vehicles in the target area.

[0073] Optionally, before the step of acquiring the collected images within the target area, road snapshot images in the target city can also be acquired; based on the road snapshot images, the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city is determined; and the target area is determined based on the number of non-motor vehicles of each non-motor vehicle manufacturer and the geographical location of the target city.

[0074] In an embodiment of the present invention, the above-mentioned road captured images can be collected by the above-mentioned drones, road cameras, satellites, mobile phones or cameras and other collection devices, and image recognition is performed based on the above-mentioned road captured images to determine the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city, and the target area is determined based on the number of non-motor vehicles of each non-motor vehicle manufacturer and the geographical location of the target city. For example, when there are too many non-motor vehicles in the prosperous commercial area of ​​the above-mentioned target city, and there are phenomena such as random parking and illegal riding, the above-mentioned prosperous commercial area can be determined as the above-mentioned target area. Or when there are fewer non-motor vehicles in the industrial area of ​​the above-mentioned target city, and the number of non-motor vehicles is often zero, the above-mentioned industrial area can be determined as the above-mentioned target area.

[0075] Optionally, in the step of determining the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on road captured images, non-motor vehicle identification can be performed based on road captured images to determine the number and color of non-motor vehicles in the target city; and the number of non-motor vehicles of each non-motor vehicle manufacturer can be determined based on the number and color of non-motor vehicles in the target city.

[0076] In an embodiment of the present invention, in general, in order to increase the recognition of non-motor vehicles, different non-motor vehicle manufacturers generally produce non-motor vehicles with different colors, patterns and sizes. Therefore, non-motor vehicle identification can be performed based on road capture images to determine the number and colors of target non-motor vehicles. Based on the above number and color, the number of non-motor vehicles of each color is determined, and the above colors are matched with non-motor vehicle models produced by all non-motor vehicle manufacturers in the above target city to determine the number of non-motor vehicles of each non-motor vehicle manufacturer.

[0077] Optionally, in the step of determining the proportion of the target detection frame in the captured image based on the target detection frame and the captured image, the width and height of the target detection frame can be determined based on the target detection frame; the width and height of the captured image can be determined based on the captured image; and the proportion of the target detection frame in the captured image can be determined based on the width and height of the target detection frame and the width and height of the captured image.

[0078] In an embodiment of the present invention, the target detection frame and the acquired image are generally rectangular, the width of the target detection frame can be determined according to the longer side of the target detection frame, the height of the target detection frame can be determined according to the shorter side of the target detection frame, the height of the acquired image can be determined according to the shorter side of the target detection frame, the width of the acquired image can be determined according to the longer side of the target detection frame, the area of ​​the target detection frame is determined based on the width and height of the target detection frame, the area of ​​the acquired image is determined based on the width and height of the acquired image, and the proportion of the target detection frame in the acquired image is determined based on the ratio of the area of ​​the target detection frame to the area of ​​the acquired image.

[0079] Specifically, the proportion of the above target detection box in the captured image can be expressed by the following formula:

[0080]

[0081] Among them, the above P i It is expressed as the proportion of the i-th picture, the above k is expressed as the total number of target detection boxes in the i-th picture, and the above j is expressed as i It is represented as the jth target detection box in the i-th image, h ji It is represented as the height of the jth target detection box in the i-th image, w ji It is represented as the width of the jth target detection box in the i-th image, h i Represented as the height of the i-th image, w i Represented as the width of the i-th image.

[0082] Optionally, in the step of determining the repeated area of ​​the target detection frame in the acquired image based on the target detection frame and the acquired image when the target detection frame is multiple target detection frames, the number of pixels in the overlapping part of the multiple target detection frames and the total number of pixels in each of the multiple target detection frames can be determined based on the target detection frame; the repeated area of ​​the target detection frame in the acquired image is determined based on the number of pixels in the overlapping part of the multiple target detection frames and the total number of pixels in each of the multiple target detection frames.

[0083] In an embodiment of the present invention, if there are two target detection frames, the number of pixels in the overlapping part between the two target detection frames can be counted, and the number of pixels of the two target detection frames can be determined based on the target detection frames and added to obtain the total number of pixels of the two target detection frames. The relative area of ​​the overlapping part can be determined based on the total number of pixels and the number of pixels in the overlapping part. Based on the number of pixels in the overlapping part and the relative area of ​​the overlapping part, the repeated area of ​​the two target detection frames in the captured image can be determined.

[0084] Specifically, the repeated area of ​​the above target detection box in the captured image can be expressed by the following formula:

[0085]

[0086] Among them, the above A i It is represented by the repeated area of ​​the target detection box in the i-th picture, k is the total number of the above collected images, and the above v i is the number of pixels in the target detection box, and the above q i1 is the number of pixels in the first target detection box, q i2 is the number of pixels of the second target detection box, u i is the smaller area in the above target detection box.

[0087] Optionally, when the target detection frame is multiple target detection frames, each target detection frame in the multiple target detection frames corresponds to a proportion, and the multiple target detection frames correspond to multiple proportions. Based on the proportions and the repeated areas, in the step of determining the non-motor vehicle scheduling strategy corresponding to the target area, the number of non-motor vehicles in the target area can be determined based on the multiple proportions and the repeated areas; the number of non-motor vehicles in the target area is compared with the preset number of non-motor vehicles in the target area to obtain a comparison result; based on the comparison result, the non-motor vehicle scheduling strategy corresponding to the target area is determined.

[0088] In an embodiment of the present invention, the preset number of non-motor vehicles in the target area can be determined based on regional data such as road width, vehicle flow, passenger flow, geographic location, OD passenger flow data, etc. in the target area. Specifically, if the preset number of non-motor vehicles in the target area is 100 and the number of non-motor vehicles in the target area is 80, the comparison result is that the preset number of non-motor vehicles in the target area is greater than the number of non-motor vehicles in the target area, and the number difference is 20. If the preset number of non-motor vehicles in the target area is 100 and the number of non-motor vehicles in the target area is 130, the comparison result is that the preset number of non-motor vehicles in the target area is less than the number of non-motor vehicles in the target area, and the number difference is 30. If the comparison result is that the preset number of non-motor vehicles in the target area is greater than the number of non-motor vehicles in the target area, and the number difference is 20, the non-motor vehicle dispatching strategy is to dispatch 20 non-motor vehicles to the target area. If the comparison result is that the preset number of non-motor vehicles in the target area is less than the number of non-motor vehicles in the target area, and the number difference is 30, the non-motor vehicle dispatching strategy is to transfer any 30 non-motor vehicles in the target area away from the target area.

[0089] Optionally, in the step of determining the number of non-motor vehicles in the target area based on multiple proportions and repeated areas, the overlapping area increment corresponding to the target area can be determined based on multiple proportions and repeated areas; the average proportion corresponding to the target area can be determined based on multiple proportions; and the number of non-motor vehicles in the target area can be determined based on the average proportion and the overlapping area increment.

[0090] In an embodiment of the present invention, the above-mentioned overlapping area increment (Avgs) can be understood as the amount of change in the overlapping area between two or more target detection frames in the process of determining the number of non-motor vehicles according to the overlapping area. Generally speaking, the calculation of the above-mentioned overlapping area increment requires at least 100 sample acquisition images of the target area to be extracted as data for calculation. The above-mentioned average proportion can be determined based on the target detection frame proportion of multiple sample acquisition images in the above-mentioned target area. Generally speaking, at least 100 sample acquisition images of the target area need to be extracted as data to calculate the average proportion of the target detection frame. The number of non-motor vehicles in the target area is determined based on the average proportion of the target detection frame in the above-mentioned sample acquisition images, the target detection frame proportion of the acquisition images, and the overlapping area increment.

[0091] Specifically, the above average proportion can be expressed by the following formula:

[0092]

[0093] The above C is the average proportion, the above k is the number of the above sample collection images, and the above k is generally 100. p1 represents the proportion of the target detection frame of the first sample collection image, and p2 represents the proportion of the target detection frame of the second sample collection image. i It is represented as the target detection box ratio of the i-th sample collection image.

[0094] The number of non-motor vehicles can be expressed by the following formula:

[0095]

[0096] Wherein, N is the number of non-motor vehicles in the above-mentioned collected images, C is the average proportion of the target detection frame obtained based on the above-mentioned sample collected images, and P i is the proportion of the target detection box in the above captured image, and Avgs is the overlap area increment.

[0097] It should be noted that the non-motor vehicle dispatching method provided in the embodiment of the present invention can be applied to devices such as photographing equipment, smart phones, computers, servers, etc. that can perform non-motor vehicle dispatching.

[0098] like Figure 2 As shown, an embodiment of the present invention provides a non-motor vehicle dispatching device, the non-motor vehicle dispatching device comprising:

[0099] The first acquisition module 201 is used to acquire an image in a target area;

[0100] A first recognition module 202 is used to recognize the non-motor vehicle in the collected image based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle;

[0101] A first determination module 203, configured to determine a proportion and a repeated area of ​​the target detection frame in the captured image based on the target detection frame and the captured image;

[0102] A second determination module 204 is used to determine a non-motor vehicle dispatching strategy corresponding to the target area based on the proportion and the repeated area;

[0103] The first dispatching module 205 is configured to dispatch non-motor vehicles to the target area based on the non-motor vehicle dispatching strategy.

[0104] Optionally, the non-motor vehicle dispatching device further includes:

[0105] The second acquisition module is used to acquire road snapshot images in the target city;

[0106] A third determination module is used to determine the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on the road captured image;

[0107] The fourth determination module is used to determine the target area based on the number of non-motor vehicles of each non-motor vehicle manufacturer and the geographical location of the target city.

[0108] Optionally, the third determining module includes:

[0109] A first determination submodule is used to identify non-motor vehicles based on the road captured image, and determine the number and color of non-motor vehicles in the target city;

[0110] The second determination submodule is used to determine the number of non-motor vehicles of each non-motor vehicle manufacturer based on the number and color of non-motor vehicles in the target city.

[0111] Optionally, the first determining module 203 includes:

[0112] A third determination submodule is used to determine the width and height of the target detection frame based on the target detection frame;

[0113] A fourth determination submodule, configured to determine a width and a height of the acquired image based on the acquired image;

[0114] The fifth determination submodule is used to determine the proportion of the target detection frame in the collected image based on the width and height of the target detection frame and the width and height of the collected image.

[0115] Optionally, the first determining module 203 further includes:

[0116] A sixth determination submodule, configured to determine, based on the target detection frame, the number of pixels in the overlapping portion of the plurality of target detection frames and the total number of pixels in each of the plurality of target detection frames;

[0117] The seventh determination submodule is used to determine the repeated area of ​​the target detection frame in the collected image based on the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each target detection frame among the multiple target detection frames.

[0118] Optionally, the second determining module 204 includes:

[0119] An eighth determination submodule, configured to determine the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas;

[0120] A first comparison submodule is used to compare the number of non-motor vehicles in the target area with the preset number of non-motor vehicles in the target area to obtain a comparison result;

[0121] The ninth determination submodule is used to determine the non-motor vehicle dispatching strategy corresponding to the target area based on the comparison result.

[0122] Optionally, the eighth determining submodule includes:

[0123] A first determining unit, configured to determine an overlap area increment corresponding to the target area based on the multiple proportions and the overlap areas;

[0124] A second determining unit, configured to determine an average proportion corresponding to the target area based on the multiple proportions;

[0125] The third determination unit is used to determine the number of non-motor vehicles in the target area based on the average proportion and the overlapping area increment.

[0126] It should be noted that the non-motor vehicle dispatching device provided in the embodiment of the present invention can be applied to devices such as shooting equipment, smart phones, computers, servers, etc. that can perform non-motor vehicle dispatching.

[0127] The non-motor vehicle dispatching device provided in the embodiment of the present invention can implement each process implemented by the non-motor vehicle dispatching method in the above method embodiment and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0128] See also Figure 3 , Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, it includes: a memory 302, a processor 301, and a computer program of a non-motor vehicle dispatching method stored in the memory 302 and executable on the processor 301, wherein:

[0129] The processor 301 is used to call the computer program stored in the memory 302 and execute the following steps:

[0130] Acquire a captured image within a target area;

[0131] Identify the non-motor vehicle in the captured image based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle;

[0132] Based on the target detection frame and the collected image, determining a proportion of the target detection frame in the collected image and a repeated area, wherein the repeated area is an area of ​​overlap between target detection frames when there are multiple target detection frames;

[0133] Based on the proportion and the repeated area, determining a non-motor vehicle dispatching strategy corresponding to the target area;

[0134] Non-motor vehicles are dispatched to the target area based on the non-motor vehicle dispatch strategy.

[0135] Optionally, before acquiring the captured image in the target area, the method executed by the processor 301 further includes:

[0136] Acquire road snapshot images in the target city;

[0137] Determining the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on the captured road image;

[0138] The target area is determined based on the number of non-motor vehicles of each non-motor vehicle manufacturer and the geographical location of the target city.

[0139] Optionally, the processor 301 determines the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on the captured road image, including:

[0140] Identify non-motor vehicles based on the captured road image to determine the number and color of non-motor vehicles in the target city;

[0141] The number of non-motor vehicles of each non-motor vehicle manufacturer is determined based on the number and color of the non-motor vehicles in the target city.

[0142] Optionally, the determining, based on the target detection frame and the captured image, a proportion of the target detection frame in the captured image, performed by the processor 301 includes:

[0143] Based on the target detection frame, determine the width and height of the target detection frame;

[0144] Based on the acquired image, determining the width and height of the acquired image;

[0145] Based on the width and height of the target detection frame and the width and height of the acquired image, the proportion of the target detection frame in the acquired image is determined.

[0146] Optionally, the target detection frame executed by the processor 301 is a plurality of target detection frames, and the determining, based on the target detection frame and the acquired image, of a repetition area of ​​the target detection frame in the acquired image includes:

[0147] Based on the target detection frames, determine the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each of the multiple target detection frames;

[0148] Based on the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each target detection frame among the multiple target detection frames, a repeated area of ​​the target detection frame in the acquired image is determined.

[0149] Optionally, the target detection frame executed by the processor 301 is a plurality of target detection frames, each of the plurality of target detection frames corresponds to a proportion, and the plurality of target detection frames correspond to a plurality of proportions; and determining the non-motor vehicle dispatching strategy corresponding to the target area based on the proportions and the repeated area includes:

[0150] Determining the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas;

[0151] Comparing the number of non-motor vehicles in the target area with the preset number of non-motor vehicles in the target area to obtain a comparison result;

[0152] Based on the comparison result, a non-motor vehicle dispatching strategy corresponding to the target area is determined.

[0153] Optionally, the determining of the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas performed by the processor 301 includes:

[0154] Based on the multiple proportions and repeated areas, determining an overlap area increment corresponding to the target area;

[0155] Based on the multiple proportions, determining an average proportion corresponding to the target area;

[0156] Based on the average proportion and the overlap area increment, the number of non-motor vehicles in the target area is determined.

[0157] It should be noted that the electronic device provided in the embodiment of the present invention can be applied to smart phones, computers, servers and other devices that can perform the non-motor vehicle dispatching method.

[0158] The electronic device provided in the embodiment of the present invention can implement each process implemented by the non-motor vehicle dispatching method in the above method embodiment and can achieve the same beneficial effect. To avoid repetition, it will not be described here.

[0159] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the non-motor vehicle scheduling method or the application-side non-motor vehicle scheduling method provided in the embodiment of the present invention are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0160] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0161] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A non-motor vehicle dispatching method, characterized in that: The method comprises the following steps: Acquire a captured image within a target area; Identify the non-motor vehicle in the captured image based on a preset target recognition algorithm to obtain a target detection frame corresponding to the non-motor vehicle; Based on the target detection frame and the collected image, determining a proportion of the target detection frame in the collected image and a repeated area, wherein the repeated area is an area of ​​overlap between target detection frames when there are multiple target detection frames; Based on the proportion and the repeated area, determining a non-motor vehicle dispatching strategy corresponding to the target area; Non-motor vehicles are dispatched to the target area based on the non-motor vehicle dispatch strategy.

2. The non-motor vehicle dispatching method according to claim 1, characterized in that: Before acquiring the captured image in the target area, the method further includes: Acquire road snapshot images in the target city; Determining the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city based on the captured road image; The target area is determined based on the number of non-motor vehicles of each non-motor vehicle manufacturer and the geographical location of the target city.

3. The non-motor vehicle dispatching method according to claim 2, characterized in that: The determining, based on the captured road image, the number of non-motor vehicles of each non-motor vehicle manufacturer in the target city includes: Identify non-motor vehicles based on the captured road image to determine the number and color of non-motor vehicles in the target city; The number of non-motor vehicles of each non-motor vehicle manufacturer is determined based on the number and color of the non-motor vehicles in the target city.

4. The non-motor vehicle dispatching method according to any one of claims 1 to 3, characterized in that: The determining, based on the target detection frame and the captured image, a proportion of the target detection frame in the captured image includes: Based on the target detection frame, determine the width and height of the target detection frame; Based on the acquired image, determining the width and height of the acquired image; Based on the width and height of the target detection frame and the width and height of the acquired image, the proportion of the target detection frame in the acquired image is determined.

5. The non-motor vehicle dispatching method according to any one of claims 1 to 4, characterized in that: The target detection frame is a plurality of target detection frames, and based on the target detection frames and the acquired image, determining the repetitive area of ​​the target detection frames in the acquired image includes: Based on the target detection frames, determine the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each of the multiple target detection frames; Based on the number of pixels in the overlapping parts of the multiple target detection frames and the total number of pixels in each target detection frame among the multiple target detection frames, a repeated area of ​​the target detection frame in the acquired image is determined.

6. The non-motor vehicle dispatching method according to any one of claims 1 to 5, characterized in that: The target detection frame is a plurality of target detection frames, each target detection frame in the plurality of target detection frames corresponds to a proportion, and the plurality of target detection frames correspond to a plurality of proportions; The determining of the non-motor vehicle dispatching strategy corresponding to the target area based on the proportion and the repeated area includes: Determining the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas; Comparing the number of non-motor vehicles in the target area with the preset number of non-motor vehicles in the target area to obtain a comparison result; Based on the comparison result, a non-motor vehicle dispatching strategy corresponding to the target area is determined.

7. The non-motor vehicle dispatching method according to claim 6, characterized in that: The determining the number of non-motor vehicles in the target area based on the multiple proportions and repeated areas includes: Based on the multiple proportions and repeated areas, determining an overlap area increment corresponding to the target area; Based on the multiple proportions, determining an average proportion corresponding to the target area; Based on the average proportion and the overlap area increment, the number of non-motor vehicles in the target area is determined.

8. A non-motor vehicle dispatching device, characterized in that: The non-motor vehicle dispatching device comprises: A first acquisition module is used to acquire a captured image within a target area; A first recognition module, configured to recognize the non-motor vehicle in the captured image based on a preset target recognition algorithm, and obtain a target detection frame corresponding to the non-motor vehicle; A first determination module, configured to determine a proportion and a repeated area of ​​the target detection frame in the captured image based on the target detection frame and the captured image; A second determination module is used to determine the non-motor vehicle dispatching strategy corresponding to the target area based on the proportion and the repeated area; The first scheduling module is used to schedule non-motor vehicles in the target area based on the non-motor vehicle scheduling strategy.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the non-motor vehicle dispatching method as claimed in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the non-motor vehicle dispatching method according to any one of claims 1 to 7 are implemented.

Images