A radar and camera fusion speed measurement method, system, device and storage medium

Abstract

The embodiment of the application provides a radar and camera fusion speed measurement method, system, device and storage medium, the method comprises the following steps: dividing the target river surface water flow into a plurality of grid shapes; collecting the river surface water flow image in the grid shape through the camera device; preprocessing the collected water flow image; calculating the water flow speed according to the preprocessed picture; and fusing and correcting the calculated water flow speed and the water flow speed detected by the radar. Through the fusion correction of the calculated water flow speed and the water flow speed detected by the radar, the weather environment can be classified by the lightweight model, the radar speed measurement and the image speed measurement are fused, the surface speed measurement is more accurate through the correction of the image speed measurement.

Description

TECHNICAL FIELD

[0001] The embodiments of the present application belong to the technical field of communication, and particularly relate to a radar and camera fusion speed measurement method, system, device and storage medium. BACKGROUND

[0002] The surface flow velocity of the river can effectively reflect the water regime. The surface flow velocity of a calm river generally remains between 0-1 m / s; when surging, the flow velocity is 1-3 m / s; when flowing turbulently, the flow velocity can reach a high flow velocity of 4-5 m / s or above. In the case of a broken flow, the flow velocity gradually decreases to less than 0.5 m / s, and eventually there is no water flow.

[0003] In addition, the maximum surface flow velocity during a flood disaster can reach 8-12 m / s. Therefore, real-time mastery of the surface hydrological change law of a natural river is not only a scientific basis for the study of river hydrology, river dynamics and the like, but also plays a crucial role in monitoring the hydrological phenomena of the river, forecasting and preventing floods. Due to the complex turbulent characteristics of the natural river and the complex on-site environment around the river, the difficulty of hydrological testing is greatly increased.

[0004] Therefore, in the detection process, through single radar Doppler speed measurement or through camera feature movement speed measurement, the traditional method, the radar speed measurement is relatively simple and the range is small, and the camera movement speed measurement is easily affected by the environment and light interference, which leads to the problem that the radar obtains the surface flow velocity through the transmission and reception of microwaves and the Doppler shift, the coverage area and the microwave attenuation limit the surface flow velocity coverage, and at the same time the surface flow velocity is uneven, and the weather factors cause the feature points to deteriorate. SUMMARY

[0005] The purpose of the embodiments of the present application is to provide a radar and camera fusion speed measurement method, system, device and storage medium, which can classify the weather environment through fusion correction of the calculated water flow velocity and the water flow velocity detected by the radar, fuse the radar speed measurement and the image speed measurement, correct the image speed measurement, and make the surface speed measurement more accurate, thereby solving the problems in the background art.

[0006] In order to solve the above technical problems, the technical scheme of the radar and camera fusion speed measurement method, system, device and storage medium provided by the embodiments of the present application is as follows:

[0007] In a first aspect, the embodiments of the present application disclose a radar and camera fusion speed measurement method, which comprises the following steps:

[0008] The camera device is installed and configured outdoors.

[0009] Dividing the target river surface flow into a plurality of grid shapes;

[0010] Collecting river surface flow images in the grid shape by a camera device;

[0011] Pretreating the collected flow images;

[0012] Calculating the flow velocity according to the pretreated images;

[0013] Fusing and correcting the calculated flow velocity with the flow velocity detected by radar.

[0014] In the preferred embodiment of any of the above solutions, the camera device is installed and configured outdoors, comprising:

[0015] Surveying the surrounding environment of the target river to be monitored;

[0016] Installing the camera at a suitable installation position. If the river has a bridge, the camera is installed on the bridge. If the river has no bridge, a column is erected on the flat river bank or a cableway is erected to install the camera. The selected camera can be adjusted to project parallel to the river surface by adjusting the angle of the ball machine, and then the focal length of the camera is adjusted so that the captured image only contains the flow image.

[0017] In the preferred embodiment of any of the above solutions, the collecting of river surface flow images in the grid shape comprises:

[0018] The camera shoots river surface flow videos under various weather conditions and at various time periods at a frame rate of 60 fps;

[0019] A flow velocity meter or a velocity radar is used to record the river flow velocity at the corresponding time of each video, and the video data is transmitted to the central monitoring and management platform through a wireless network.

[0020] In the preferred embodiment of any of the above solutions, the collecting of river surface flow images in the grid shape further comprises:

[0021] Using video processing software to frame-by-frame intercept flow images;

[0022] Flow images in the same flow velocity range are classified into the same category label, and their flow velocity values are set as the flow velocity range;

[0023] According to the mapping relationship between the collected images and the category labels, a training sample data set is established;

[0024] According to the corresponding relationship between the category labels and the flow velocity range, a relationship mapping table of the category labels and the flow velocity is established.

[0025] In any of the above-mentioned solutions, in a preferred embodiment, the pre-processing of the captured water flow image comprises:

[0026] The intercepted water flow image is subjected to grayscale, histogram equalization, contrast enhancement and dimension reduction preprocessing.

[0027] The pre-processed image is converted into a computer vector form for storage.

[0028] In any of the above-mentioned solutions, in a preferred embodiment, the resolution of the water flow image is 1920 pixels x 1080 pixels.

[0029] In any of the above-mentioned solutions, in a preferred embodiment, the fusion correction of the calculated water flow velocity and the water flow velocity detected by the radar comprises:

[0030] The water flow velocity V a inside the grid detected by the radar is calculated.

[0031] The water flow velocity V b inside the grid detected by the camera is calculated.

[0032] The fusion water flow velocity V c is calculated, and V c = V a x p + V b x (1-p), wherein p is a velocity weight, and 0≤p<1, and is related to environmental factors, such as day, night, sunny day, cloudy day, when in the day or sunny day, the value of p is greater than in the night or cloudy day.

[0033] Compared with the prior art, the radar and camera fusion speed measurement method according to the embodiments of the present application can classify the weather environment through the fusion correction of the calculated water flow velocity and the water flow velocity detected by the radar, so as to fuse the radar speed measurement and the image speed measurement, and through the correction of the image speed measurement, the surface speed measurement is more accurate.

[0034] In a second aspect, a radar and camera fusion speed measurement system comprises:

[0035] The installation module is configured to install and configure the camera device outdoors, which comprises surveying the surrounding environment of the target river to be monitored, installing the camera on a suitable installation position, if the river has a bridge, installing the camera on the bridge, if the river has no bridge, selecting a flat river bank to erect a column or install a cableway to install the camera, and adjusting the angle of the camera to make the camera project parallel to the river surface, and then adjusting the focal length of the camera to make the captured image only contain the water flow image.

[0036] The collection module is used for collecting the river surface water flow image in the grid by the camera device, and the collection of the river surface water flow image in the grid comprises: the camera shoots the river surface water flow video under various time periods and various weather conditions at a frame rate of 60 fps; the flow velocity of the river at the moment corresponding to each video is recorded by using a flow velocity meter or a velocity radar, the video data is transmitted to the central monitoring management platform through a wireless network, and the water flow image is intercepted frame by frame by using a video processing software; the water flow images in the same flow velocity range are classified into the same category label, and the flow velocity value of the water flow images is set as the flow velocity range; a training sample data set is established according to the mapping relationship of the collected images and the category label; and a relationship mapping table of the category label and the flow velocity is established according to the corresponding relationship of the category label and the flow velocity range;

[0037] The processing module is used for pre-processing the collected water flow image, and the pre-processing of the collected water flow image comprises: the intercepted water flow image is subjected to gray processing, histogram equalization, contrast enhancement and dimension reduction preprocessing; and the pre-processed image is converted into a computer vector form for storage.

[0038] The calculation module is used for calculating the water flow velocity according to the pre-processed picture.

[0039] The correction module is used for fusing and correcting the calculated water flow velocity and the water flow velocity detected by the radar, and the fusing and correcting of the calculated water flow velocity and the water flow velocity detected by the radar comprises: the water flow velocity V a ,

[0040] The water flow velocity V b detected by the camera in the grid is calculated.

[0041] The fused water flow velocity V c is calculated, and V c = V a × p + V b × (1-p), wherein p is a velocity weight, and 0≤p<1, and is related to environmental factors, and the environmental factors are day, night, sunny day and cloudy day, and the value of p is greater when it is day or a sunny day than when it is night or a cloudy day.

[0042] The second aspect has the same beneficial effects as the first aspect, and therefore will not be described here.

[0043] In a third aspect, a radar and camera fusion speed measurement device is provided, comprising:

[0044] A memory is configured to store a computer program.

[0045] A processor is configured to execute the computer program to implement the steps of the radar and camera fusion speed measurement method.

[0046] The beneficial effects of the third aspect are consistent with those of the first aspect, so they will not be described in detail here.

[0047] In a fourth aspect, a storage medium stores a computer program thereon, which, when executed by a processor, implements the speed measurement method based on radar and camera fusion.

[0048] The beneficial effects of the fourth aspect are consistent with those of the first aspect, so they will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings described herein are used to provide a further understanding of the present application and constitute a component of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. Some specific embodiments of the present application will be described in detail in an illustrative and non-restrictive manner with reference to the drawings. The same reference numerals in the drawings indicate the same or similar components or components. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings:

[0050] Figure 1 This is a flow chart of the speed measurement method based on radar and camera fusion in an embodiment of the present application.

[0051] Figure 2 This is a schematic diagram of a speed measurement system based on radar and camera fusion according to an embodiment of the present application.

[0052] Figure 3 This is a schematic diagram of a speed measurement device based on the fusion of radar and camera in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only embodiments of a component of the present invention, not embodiments of the entire component. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0054] Example

[0055] like Figure 1 As shown, the embodiment of the present application provides a speed measurement method based on radar and camera fusion, the method comprising the following steps:

[0056] Install and configure the camera equipment outdoors;

[0057] Divide the surface water flow of the target river into multiple grids;

[0058] collecting the river surface flow image in the grid by the camera;

[0059] preprocessing the collected flow image;

[0060] calculating the flow velocity according to the preprocessed image;

[0061] fusing and correcting the calculated flow velocity and the flow velocity detected by the radar.

[0062] In the radar and camera fusion speed measurement method described in the embodiment of the present application, by fusing and correcting the calculated flow velocity and the flow velocity detected by the radar, the weather environment can be classified by a lightweight model, the radar speed measurement and image speed measurement are fused, and the surface speed measurement is more accurate by correcting the image speed measurement.

[0063] In the present application, the camera is installed and configured outdoors, comprising:

[0064] surveying the surrounding environment of the monitoring target river;

[0065] installing the camera at a suitable installation position, if the river has a bridge, installing the camera on the bridge, if the river has no bridge, selecting a flat river bank to erect a column or install a cableway to install the camera, the selected camera can be adjusted to be parallel to the river surface by adjusting the angle of the ball machine, and then adjusting the focal length of the camera to capture only the flow image; the river surface flow image in the grid is collected, comprising:

[0066] The camera shoots the river surface flow video under various weather conditions at a frame rate of 60fps;

[0067] The flow velocity or speed measurement radar is used to record the river flow velocity at the corresponding time of each video, and the video data is transmitted to the center monitoring management platform through the wireless network; the river surface flow image in the grid is collected, further comprising:

[0068] The flow image is frame by frame intercepted by using the video processing software;

[0069] The flow images in the same flow velocity range are classified into the same category label, and their flow velocity values are set as the flow velocity range;

[0070] According to the mapping relationship between the collected images and the category labels, a training sample data set is established;

[0071] According to the corresponding relationship between the category labels and the flow velocity range, a relationship mapping table of the category labels and the flow velocity is established.

[0072] In the radar and camera fusion speed measurement method described in the embodiment of the present application, the greater the angle between the camera and the water surface, that is, the more the camera is inclined, the more the captured picture is cluttered, and the less clear the extracted water flow features are; the smaller the angle, the closer the camera is to parallel to the water surface, the clearer the captured water flow picture is, and the more obvious the extracted water flow features are. Therefore, if the environment around the river allows, the camera should try to vertically capture the image. In special cases, the angle should also be reduced as much as possible.

[0073] In the present application, the pre-processing of the collected water flow image comprises:

[0074] The intercepted water flow image is subjected to grayscale, histogram equalization, contrast enhancement and dimension reduction preprocessing.

[0075] The pre-processed image is converted into a computer vector form for storage, and the resolution of the water flow image is 1920 pixels x 1080 pixels.

[0076] In the present application, the fusion correction of the calculated water flow speed and the radar detected water flow speed comprises:

[0077] The river surface water flow speed V in the grid detected by the radar is calculated a ,

[0078] The river surface water flow speed V in the grid detected by the camera is calculated b ;

[0079] The fusion water flow speed V is calculated c , and V c = V a x p + V b x (1-p), wherein p is the speed weight, and 0≤p<1, and is related to environmental factors, such as day, night, sunny day, cloudy day, when in the day or sunny day, the value of p is greater than in the night or cloudy day.

[0080] In the radar and camera fusion speed measurement method described in the embodiment of the present application, the intercepted water flow image is subjected to grayscale, histogram equalization, contrast enhancement, dimension reduction and other preprocessing. The grayscale process of converting a color image into a grayscale image can reduce the amount of calculation in subsequent processing. Histogram equalization can effectively enhance the local contrast of the image, especially for water flow images, which can better highlight the features. Through contrast enhancement, the differences between different object features in the image can be expanded, and useless information can be suppressed to improve the recognition rate. By dimension reduction, redundant information can be effectively removed, useful features can be extracted, and the recognition efficiency can be improved. Finally, the pre-processed image is converted into a computer vector form for storage.

[0081] Compared with the prior art, the radar and camera fusion speed measurement method of the embodiment of the application can classify the weather environment through the fusion correction of the calculated water flow speed and the water flow speed detected by the radar, fuse the radar speed measurement and the image speed measurement, and correct the image speed measurement to make the surface speed measurement more accurate.

[0082] As shown in Figure 2 the second aspect, a radar and camera fusion speed measurement system comprises:

[0083] The installation module is configured to install and configure the camera device outdoors, and the installation and configuration of the camera device outdoors comprises: surveying the surrounding environment of the target river to be monitored; installing the camera on a suitable installation position, if the river has a bridge, installing the camera on the bridge, if the river has no bridge, selecting a flat river bank to erect a column or install a cableway to install the camera, and adjusting the angle of the camera to make the camera project on the river surface in parallel, and then adjusting the focal length of the camera to make the captured image only exist the water flow image;

[0084] The acquisition module is configured to acquire the river surface water flow image in the grid through the camera device, and the acquisition of the river surface water flow image in the grid comprises: shooting the river surface water flow video under various weather conditions in each time period at a frame rate of 60 fps; recording the river flow speed at the corresponding time of each video using a flowmeter or a speed measurement radar, transmitting the video data to the central monitoring and management platform through a wireless network, and using video processing software to intercept the water flow image frame by frame; classifying the water flow images in the same flow speed range into the same category label, and setting their flow speed values as the flow speed range; establishing a training sample data set according to the mapping relationship between the acquired images and the category labels; and establishing a relationship mapping table between the category labels and the flow speed according to the corresponding relationship between the category labels and the flow speed range;

[0085] The processing module is configured to pre-process the acquired water flow image, and the pre-processing of the acquired water flow image comprises: performing gray-scale and dimension reduction preprocessing on the intercepted water flow image; and converting the pre-processed image into a computer vector form for storage;

[0086] The calculation module is configured to calculate the water flow speed according to the pre-processed image.

[0087] The correction module is configured to fuse and correct the calculated water flow speed and the water flow speed detected by the radar, and the fusion and correction of the calculated water flow speed and the water flow speed detected by the radar comprises: calculating the water flow speed V a ,

[0088] calculating the water flow speed Vb ;

[0089] Calculate the fused water velocity V c , and V c =V a ×p+V b ×(1-p), where p is the velocity weight, and 0≤p<1, and is related to environmental factors, such as daytime, nighttime, sunny, and cloudy. During daytime and sunny days, image acquisition is more accurate, so the p value is small. For example, at night and sunny days, the image acquisition information is not as accurate as the radar acquisition information, so the p value will be larger. For example, the p value is 0.55, so the water velocity V after fusion is c =V a ×0.55+V b ×(1-0.55). For another example, during the daytime or on cloudy days, the accuracy of image acquisition information is higher than that of radar acquisition information. At this time, the p value is small, such as 0.4, so the water velocity V after fusion is c =V a ×0.4+V b ×(1-0.4), by calculating the fusion water flow velocity according to the influence of different external environments, the detection can be made more accurate. Among them, the environmental factors are inferred through deep learning. For example, at night or on rainy days, the accuracy of image acquisition is lower than that of radar acquisition, so at this time, the p value is relatively large. For example, if it is 0.6, the water flow velocity after fusion is V c =V a ×0.6+V b ×(1-0.6), for example, when it is daytime and rainy, the accuracy of radar acquisition and image acquisition is similar, so the p value is 0.5, so the water velocity V after fusion c =V a ×0.5+V b ×(1-0.5), for example, when it is at night and sunny, the accuracy of radar acquisition is greater than the accuracy of image acquisition, so the p value is 0.55, so the water velocity V after fusion c =V a ×0.55+V b ×(1-0.55); For another example, when it is daytime and cloudy, the accuracy of image acquisition is greater than the accuracy of radar acquisition, so the p value is 0.45, so the water velocity V after fusion c =V a ×0.45+V b×(1-0.45); For another example, at night, when it is cloudy, the accuracy of image acquisition is similar to that of radar acquisition, so the p value is 0.5, so the water velocity V after fusion is c =V a ×0.5+V b ×(1-0.5).

[0090] Thirdly, as Figure 3 As shown, a speed measurement device based on radar and camera fusion includes: a memory for storing a computer program; a processor for implementing the steps of the speed measurement method based on radar and camera fusion when executing the computer program.

[0091] The processor is configured to control overall operations of the measurement device to complete all or part of the steps in the above radar and camera fusion speed measurement method. The memory is configured to store various types of data to support the operation of the measurement device, which can include, for example, instructions for any application or method operating on the measurement device, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component can include a screen and an audio component. The screen can be a touch screen, for example, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory or transmitted through the communication component. The audio component also includes at least one speaker configured to output audio signals. The I / O interface provides an interface between the processor and other interface modules, which can be a keyboard, a mouse, a button, and the like. These buttons can be virtual buttons or physical buttons. The communication component is configured to perform wired or wireless communication between the measurement device and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, 4G or 5G, or a combination of one or more of them, so the corresponding communication component can include a Wi-Fi module, a Bluetooth module, an NFC module.

[0092] In an exemplary embodiment, the measurement device can be implemented by one or more Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic elements for executing the above-mentioned radar and camera fusion speed measurement method.

[0093] In another exemplary embodiment, a computer readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the above-mentioned radar and camera fusion speed measurement method. For example, the computer readable storage medium can be the above-mentioned memory including program instructions, which can be executed by the processor of the measurement device to complete the above-mentioned radar and camera fusion speed measurement method.

[0094] Corresponding to the above method embodiments, the embodiments of the present disclosure also provide a readable storage medium, which can be mutually corresponding to the above-mentioned radar and camera fusion speed measurement method as described below.

[0095] In a fourth aspect, a readable storage medium is provided, and the readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned radar and camera fusion speed measurement method.

[0096] The readable storage medium can be specifically a U disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various readable storage media that can store program codes.

[0097] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the above embodiments, or make equivalent replacements to some or all components of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A speed measurement method based on radar and camera fusion, characterized in that: The method comprises the following steps: Install and configure the camera equipment outdoors; Divide the surface water flow of the target river into multiple grids; The image of the river surface water flow in the grid is collected by a camera device; Preprocessing the collected water flow images; Calculate the water flow velocity based on the preprocessed images; The calculated water flow velocity is fused and corrected with the water flow velocity detected by the radar; The step of fusing and correcting the calculated water flow velocity with the water flow velocity detected by the radar includes: Calculate the surface water velocity V of the river within the grid detected by the radar a , Calculate the surface water velocity V of the river within the grid detected by the camera b ; Calculate the fused water velocity V c , and V c =V a ×p+V b ×(1-p), where p is the speed weight, and 0≤p<1, and is related to environmental factors, wherein the environmental factors are daytime, nighttime, sunny, and cloudy. When it is daytime or sunny, the p value is greater than that at night or cloudy. The environmental factors are obtained through deep learning reasoning.

2. The speed measurement method based on radar and camera fusion according to claim 1 is characterized in that: The outdoor installation and configuration of the camera device includes: Survey the surrounding environment of the target river to be monitored; Install the camera in a suitable location. If there is a bridge over the river, install the camera on the bridge. If there is no bridge over the river, choose a flat river bank to erect pillars or set up a cableway to install the camera. The selected camera can be projected parallel to the surface of the river by adjusting the angle of the dome camera, and then adjust the focal length of the camera so that the captured image only contains the image of the water flow.

3. The speed measurement method based on radar and camera fusion according to claim 1 or 2, characterized in that: The collecting of river surface water flow images in the grid includes: The camera captures river surface flow videos at a frame rate of 60fps at various time periods and weather conditions; Use a flow meter or speed radar to record the river flow speed at the corresponding moment of each video, and transmit the video data to the central monitoring and management platform via a wireless network.

4. The speed measurement method based on radar and camera fusion according to claim 3 is characterized in that: The collecting of river surface water flow images in the grid also includes: Use video processing software to capture water flow images frame by frame; Classify water flow images within the same flow rate range into the same category label and set their flow rate values ​​to that flow rate range; According to the mapping relationship between the collected images and category labels, a training sample data set is established; According to the correspondence between category labels and flow rate ranges, a relationship mapping table between category labels and flow rates is established.

5. The speed measurement method based on radar and camera fusion according to claim 4 is characterized in that: The preprocessing of the collected water flow image includes: The intercepted water flow image is preprocessed by grayscale conversion, histogram equalization, contrast enhancement, and dimensionality reduction; The pre-processed image is converted into computer vector form for storage.

6. The speed measurement method based on radar and camera fusion according to claim 5 is characterized in that: The resolution of the water flow image is 1920 pixels×1080 pixels.

7. A speed measurement system based on radar and camera fusion, characterized in that: include: An installation module is used to install and configure the camera device outdoors to divide the surface water flow of the target river into multiple grids; The outdoor installation and configuration of the camera device includes: surveying the surrounding environment of the target river to be monitored; installing the camera at a suitable installation location, if the river has a bridge, then installing the camera on the bridge; if the river has no bridge, then selecting a flat river bank to erect pillars or set up a cableway to install the camera, and adjusting the angle of the dome camera so that the camera is projected parallel to the surface of the river, and then adjusting the focal length of the camera so that the image captured only contains the image of the water flow; The acquisition module is used to acquire river surface water flow images within a grid pattern using a camera device, wherein the acquisition of river surface water flow images within the grid pattern includes: using a camera to shoot river surface water flow videos at a frame rate of 60fps in various time periods and under various weather conditions; using a current meter or speed radar to record the river flow velocity at the corresponding moment of each video segment, transmitting the video data to a central monitoring and management platform via a wireless network, and using video processing software to capture water flow images frame by frame; classifying water flow images within the same flow velocity range into the same category label, and setting their flow velocity values ​​to the flow velocity range; establishing a training sample data set based on the mapping relationship between the acquired images and the category labels; and establishing a relationship mapping table between category labels and flow velocity based on the corresponding relationship between category labels and flow velocity ranges. The processing module is used to pre-process the collected water flow image, wherein the pre-processing includes: gray-scaling, histogram equalization, contrast enhancement, and dimensionality reduction pre-processing of the intercepted water flow image; and converting the pre-processed image into a computer vector format for storage; A calculation module is used to calculate the water flow velocity based on the pre-processed image; The correction module is used to fuse the calculated water flow velocity with the water flow velocity detected by the radar, and the fusion correction of the calculated water flow velocity with the water flow velocity detected by the radar includes: calculating the surface water flow velocity V of the river in the grid detected by the radar a , Calculate the surface water velocity V of the river within the grid detected by the camera b ; Calculate the fused water velocity V c , and V c =V a ×p+V b ×(1-p), where p is the speed weight, and 0≤p<1, and is related to environmental factors, wherein the environmental factors are daytime, nighttime, sunny, and cloudy. When it is daytime or sunny, the p value is greater than that at night or cloudy. The environmental factors are obtained through deep learning reasoning.

8. A speed measurement device based on radar and camera fusion, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the speed measurement method based on radar and camera fusion as described in any one of claims 1 to 6 when executing the computer program.

9. A storage medium, characterized in that: A computer program is stored thereon, characterized in that when the program is executed by a processor, the speed measurement method based on radar and camera fusion as described in any one of claims 1 to 6 is implemented.

Images