Vehicle speed matching method based on underbody monitoring and underbody image stitching method

Abstract

The present application provides a vehicle speed matching method based on under-vehicle monitoring, including: Step 1, preset the line scanning width x of the line scan camera and the pixel height y of a single frame; Step 2, measure the actual maximum illumination field of view width X of the line scan camera and calculate the actual length Y corresponding to a single frame of picture; Step 3, detect the speed V of the vehicle and calculate the time T required for the vehicle to pass through the actual length Y corresponding to a single frame of picture; Step 4, automatically match the line frame Pv of the line scan camera according to the time T required for the vehicle to pass through the actual length Y corresponding to a single frame of picture; The present application also provides a method for stitching under-vehicle images; The vehicle speed matching method based on under-vehicle monitoring of the present application detects the speed V of the vehicle and automatically matches the line frame of the line scan camera to adapt to the speeds of different vehicles when passing through the lane where the line scan camera is buried, so that the line scan camera can obtain continuous and clear single-frame under-vehicle images based on the optimal line frame required for image stitching, thereby improving the quality of under-vehicle image stitching.

Description

Technical Field

[0001] This application belongs to the technical field of vehicle security inspection. More specifically, it relates to a vehicle speed matching method and a vehicle bottom image stitching method based on vehicle bottom monitoring. Background Art

[0002] In important places such as border control and customs, it is necessary to strictly control the entry and exit safety of personnel, drivers and passengers. Therefore, vehicle bottom monitoring is installed in relatively important places to prevent prohibited items or unauthorized personnel from entering the controlled area by hiding people or objects under the vehicle. Vehicle bottom monitoring equipment has been widely used in intelligent transportation, border control and customs checkpoint detection, and unmanned self-service channel traffic is also becoming more and more developed.

[0003] In the prior art, for important places that need to detect the vehicle bottom situation 24 hours a day, cameras are buried underground in advance and the camera parameters are set in advance. When a vehicle passes through the vehicle bottom monitoring lane, it is required that the vehicle travels in a straight line at a specified speed. The principle of vehicle bottom monitoring is to use a camera to scan the vehicle bottom of passing vehicles, and then perform image stitching on the scan to form a static picture for vehicle inspection personnel to detect. However, in reality, when a vehicle passes through the vehicle bottom monitoring lane, it is very difficult for the vehicle speed to accurately reach the specified passing speed, and there is generally a certain error, resulting in an increased deformation of the static picture obtained by image stitching, poor image quality, and inconsistency with the real vehicle bottom characteristics. Summary of the Invention

[0004] The purpose of the embodiments of this application is to provide a vehicle speed matching method and a vehicle bottom image stitching method based on vehicle bottom monitoring to solve the technical problem of insufficient quality of the vehicle bottom image stitching diagram in the existing vehicle bottom monitoring process.

[0005] To achieve the above object, the technical solution adopted by this application is: to provide a vehicle speed matching method based on vehicle bottom monitoring, including:

[0006] Step 1, preset the line scan width x of the line scan camera and the pixel height y of a single frame;

[0007] Step 2, measure the actual maximum irradiation field width X of the line scan camera, and calculate the actual length Y corresponding to a single frame of the picture;

[0008] Step 3, detect the speed V of the vehicle, and calculate the time T required for the vehicle to pass through the actual length Y corresponding to a single frame of the picture;

[0009] Step 4, automatically match the line frame P of the line scan camera according to the time T required for the vehicle to pass through the actual length Y corresponding to a single frame of the picture v .

[0010] Preferably, in Step 2, the method for measuring the actual maximum irradiation field width X of the line scan camera includes the steps:

[0011] Place the line scan camera after lens matching in the lane;

[0012] Place the calibration plate parallel to the ground directly above the line scan camera;

[0013] Move the calibration plate upward until the line scan camera obtains the maximum illumination field of view;

[0014] Measure the actual maximum illumination field of view width X on the calibration plate corresponding to the maximum illumination field of view obtained by the line scan camera.

[0015] Preferably, in step two, the method for calculating the actual length Y corresponding to a single-frame image includes the steps of:

[0016] Calculate the true distance s corresponding to each pixel, where:

[0017]

[0018] According to the true distance s corresponding to each pixel, calculate the actual length Y corresponding to a single-frame image, where:

[0019] Y = s·y.

[0020] Preferably, in step three, the method for detecting the speed V of the vehicle includes the steps of:

[0021] Set up inductive loops A1 and A2 at the entrance of the lane;

[0022] Pre-measure the distance S between inductive loops A1 and A2;

[0023] Sense the time interval t when the vehicle triggers inductive loops A1 and A2;

[0024] Calculate the speed V of the vehicle, where:

[0025]

[0026] Preferably, in step three, the speed V of the vehicle is detected in real time, and the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image is calculated in real time;

[0027] In step four, the line frame P of the line scan camera is automatically matched in real time according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image v .

[0028] Preferably, in step four, the line frame P of the line scan camera is automatically matched according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image v , where:

[0029]

[0030] The present application also provides a method for stitching vehicle bottom images, and the method for stitching vehicle bottom images includes the steps of:

[0031] Automatically matching the line frame P of the line scan camera by using the vehicle speed matching method based on vehicle bottom monitoring as described above v ;

[0032] The line scan camera obtains continuous single-frame vehicle bottom pictures;

[0033] Stitch the continuous single-frame vehicle bottom pictures;

[0034] Output a complete vehicle bottom monitoring scan picture.

[0035] Preferably, when the line scan camera obtains continuous single-frame vehicle bottom pictures, the fill lights on both sides of the line scan camera are turned on simultaneously.

[0036] Preferably, the method for stitching the continuous single-frame vehicle bottom pictures includes the steps of:

[0037] Detect the edge contour of the foreground object in the single-frame picture;

[0038] Extract the center line of the edge contour of the foreground object;

[0039] Take the center line as a feature of the single-frame vehicle bottom picture, and sequentially stitch the continuous single-frame vehicle bottom pictures.

[0040] Preferably, the method for sequentially stitching the continuous single-frame vehicle bottom pictures further includes the steps of:

[0041] Cut the overlapping area of the single-frame picture, find the stitching line with the minimum energy in the overlapping area, and perform fusion processing.

[0042] Compared with the prior art, the vehicle speed matching method based on vehicle bottom monitoring provided by the present application can accurately calculate the actual length Y corresponding to a single-frame picture by measuring the actual maximum illumination field width X of the line scan camera; detect the vehicle speed V, calculate the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame picture, and automatically match the line frame of the line scan camera to adapt to the speeds of different vehicles when passing through the lane where the line scan camera is buried, so that the line scan camera obtains continuous and clear single-frame vehicle bottom images based on the optimal line frame required for image stitching, thereby improving the quality of vehicle bottom image stitching.

[0043] Compared with the prior art, the vehicle bottom image stitching method provided by the present application can enable the line scan camera to obtain continuous and clear single-frame vehicle bottom images based on the optimal line frame required for image stitching by automatically matching the line frame Pv of the line scan camera by using the vehicle speed matching method based on vehicle bottom monitoring, which not only reduces the stitching difficulty but also improves the quality of vehicle bottom image stitching. Description of the Drawings

[0044] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for use in the embodiments or the description of the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0045] Figure 1 It is a schematic flowchart of the vehicle speed matching method based on underbody monitoring provided by the embodiments of the present application;

[0046] Figure 2 It is a schematic diagram of the scenario when the vehicle is performing underbody monitoring provided by the embodiments of the present application;

[0047] Figure 3 It is a schematic flowchart of the underbody image stitching method provided by the embodiments of the present application;

[0048] Figure 4 Based on Figure 3 It is a schematic diagram of consecutive single-frame underbody pictures obtained by the underbody image stitching method in the case where the steering wheel is not kept stable;

[0049] Figure 5 Based on Figure 4 It is a schematic diagram of the processing process of sequentially stitching consecutive single-frame underbody pictures with the center line as the feature;

[0050] Figure 6 Based on Figure 5 It is a schematic diagram of the effect of the underbody scan image after stitching is completed. Detailed implementation manners

[0051] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] It should be noted that when an element is referred to as "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element.

[0053] It should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present application.

[0054] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, "a plurality of" means two or more unless otherwise specifically defined.

[0055] Please refer to Figures 1 to 2 , and a vehicle speed matching method based on under-vehicle monitoring provided by an embodiment of the present application will be described below. The vehicle speed matching method based on under-vehicle monitoring includes:

[0056] Step 1, preset the line scan width x of the line scan camera and the pixel height y of a single frame.

[0057] Step 2, measure the actual maximum illumination field width X of the line scan camera and calculate the actual length Y corresponding to a single-frame picture.

[0058] Step 3, detect the speed V of the vehicle and calculate the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame picture.

[0059] Step 4, automatically match the line frame P of the line scan camera according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame picture v .

[0060] It can be understood that Step 1 and Step 2 belong to preparatory work. In Step 1, that is, for the line scan camera buried in the lane, first preset the line scan width x of the line scan camera and the pixel height y of a single frame. The values of the line scan width x and the pixel height y of a single frame need to be set according to actual situations such as the lane width, the chassis width of the vehicle, and the requirements for vehicle inspection clarity. Of course, if it is a line scan camera with fixed parameters that does not support custom settings, it can be understood as obtaining the line scan width x of the line scan camera and the pixel height y of a single frame. These belong to the performance parameters of the line scan camera and are also known parameters. If it is a 2K line scan camera, the line scan width is 2048 px. If it is a 4K line scan camera, the line scan width is 4096 px.

[0061] In step two, the actual maximum illumination field of view width X of the line scan camera refers to the actual maximum distance range illuminated by the line scan camera in cooperation with the lens after the line scan camera is buried underground, which can also be measured in advance with a ruler. In the line scan camera, since the actual maximum illumination field of view width X is much larger than the actual length Y corresponding to a single-frame image, therefore, by combining the actual maximum illumination field of view width X with the preset line scan width x of the line scan camera in step one, the true distance s corresponding to each pixel can be calculated. By combining the true distance s corresponding to each pixel with the single-frame line height pixel y, the actual distance Y corresponding to the single-frame image can be deduced. In this way, even if there are inevitable errors due to the defects of the measuring tool in the actual length Y corresponding to the single-frame image calculated by measuring the actual maximum illumination field of view width X of the line scan camera, when it is converted to the true distance s corresponding to each pixel, it can be ignored. For example, in a 4K line scan camera, the accuracy of the measuring tool is 0.01 mm, then when it is converted to the true distance s corresponding to each pixel, the maximum error is only 0.0000024 mm. In this way, the actual length Y corresponding to the single-frame image can be accurately calculated.

[0062] It can be understood that steps three and four belong to the application work. In step three, detecting the speed V of the vehicle should refer to the speed of the vehicle when it passes through the lane where the line scan camera is buried. By combining the speed V of the vehicle with the actual length Y corresponding to the single-frame image calculated in step two, the time T required for the vehicle to pass through the actual length Y corresponding to the single-frame image can be calculated.

[0063] In step four, automatically match the line frame P of the line scan camera according to the time T required for the vehicle to pass through the actual length Y corresponding to the single-frame image v to adapt to the speeds of different vehicles when passing through the lane where the line scan camera is buried, so that the line scan camera can obtain continuous and clear single-frame vehicle bottom images based on the optimal line frame required for image stitching, thus improving the quality of vehicle bottom image stitching.

[0064] It is worth supplementing that in the actual scenario application, the vehicle bottom monitoring fixed camera is composed of a line scan camera and two side fill lights. Since the light input of the line scan camera is affected by various factors such as exposure, lens, and line frequency, it is necessary to use two side fill lights to illuminate the bottom of the scanned vehicle. After the vehicle moves above it, the line scan camera will stitch the chassis images of the passing vehicles at the matched line frequency to form a complete and bright vehicle bottom monitoring scan picture.

[0065] The vehicle speed matching method based on vehicle bottom monitoring provided by this application, compared with the prior art, can accurately calculate the actual length Y corresponding to a single-frame image by measuring the actual maximum illumination field of view width X of the line scan camera; detect the speed V of the vehicle, calculate the time T required for the vehicle to pass through the actual length Y corresponding to the single-frame image, and automatically match the line frame P of the line scan camerav to adapt to the speeds of different vehicles when passing through the lane where the line scan camera is buried, so that the line scan camera can obtain continuous and clear single-frame underbody images based on the optimal line frames required for image stitching, thereby improving the quality of underbody image stitching.

[0066] In another embodiment of the present application, in step two, the method for measuring the actual maximum irradiation field of view width X of the line scan camera includes the steps of:

[0067] Bury the line scan camera after lens matching in the lane;

[0068] Place the calibration plate parallel to the ground directly above the line scan camera;

[0069] Translate the calibration plate upward until the line scan camera obtains the maximum irradiation field of view;

[0070] Measure the actual maximum irradiation field of view width X corresponding to the maximum irradiation field of view obtained by the line scan camera on the calibration plate.

[0071] It can be understood that the calibration plate can be a traditional chessboard pattern or a solid circle array pattern, or a long ruler pattern with scales, aiming to facilitate marking the corresponding distance on the calibration plate based on the maximum irradiation field of view obtained by the line scan camera after lens matching, and then measuring the actual maximum irradiation field of view width X.

[0072] In another embodiment of the present application, in step two, the method for calculating the actual length Y corresponding to a single-frame picture includes the steps of:

[0073] Calculate the true distance s corresponding to each pixel, where:

[0074]

[0075] According to the true distance s corresponding to each pixel, calculate the actual length Y corresponding to the single-frame picture, where:

[0076] Y = s·y.

[0077] It can be understood that calculating the actual length Y corresponding to the single-frame picture according to the true distance s corresponding to each pixel solves the technical problems of the too small actual length Y corresponding to the single-frame picture and large measurement errors.

[0078] In another embodiment of the present application, in step three, the method for detecting the speed V of the vehicle includes the steps of:

[0079] Set inductive sensors A1 and A2 at the entrance of the lane;

[0080] Pre-measure the distance S between inductive sensors A1 and A2;

[0081] The time interval t between the induction of the vehicle triggering the ground sensors A1 and A2;

[0082] Calculate the speed V of the vehicle, where:

[0083]

[0084] It can be understood that the ground sensors A1 and A2 can be set in the middle of the lane. The vehicle first passes through the ground sensors A1 and A2 at a constant speed, and the speed V of the vehicle is calculated to automatically match the line scan camera's line frame P according to the vehicle speed. v .

[0085] In another embodiment of the present application, in step three, the speed V of the vehicle is detected in real time, and the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image is calculated in real time;

[0086] In step four, the line scan camera's line frame P is automatically matched in real time according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image. v .

[0087] It can be understood that by detecting the speed V of the vehicle in real time, it can be ensured that even when the vehicle is accelerating or decelerating, the line scan camera's line frame P can still be automatically matched in real time according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image. v , so as to avoid the influence of frame loss and image stacking problems caused by vehicle speed changes on image stitching. For example, the speed V of the vehicle is detected in real time by a radar Doppler detector.

[0088] In another embodiment of the present application, in step four, the line scan camera's line frame P is automatically matched according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image. v , where:

[0089]

[0090] It can be understood that since within the same time, after the line frequency of stitching matches the vehicle speed at this time, the actual number of lines stitched by the line scan camera converted to the actual distance is equal to the actual distance passed by the vehicle. Thus, it can be inferred that: the time required for the line scan camera to scan a single-frame image with the line frame P v is equal to the time required for the vehicle to pass through the actual length Y corresponding to a single-frame image. That is:

[0091]

[0092] Thus, it can be inferred that

[0093] Please refer to Figure 3, this application also provides a method for stitching vehicle bottom images. The method for stitching vehicle bottom images includes the steps:

[0094] Automatically match the line scan camera's row frame P using the vehicle speed matching method based on vehicle bottom monitoring as described above v ;

[0095] The line scan camera obtains continuous single-frame images of the vehicle bottom;

[0096] Stitch the continuous single-frame images of the vehicle bottom;

[0097] Output a complete vehicle bottom monitoring scan image.

[0098] It can be understood that in actual scenario applications, the vehicle bottom monitoring fixed camera is composed of a line scan camera and two side fill lights. Since the light intake of the line scan camera is affected by various factors such as exposure, lens, and line frequency, it is necessary to use the two side fill lights to illuminate the bottom of the vehicle being scanned. That is, when the line scan camera obtains continuous single-frame images of the vehicle bottom, the two side fill lights of the line scan camera are turned on simultaneously. After the vehicle moves above it, the line scan camera will stitch the chassis images of the passing vehicle at the matched line frequency to form a complete and bright vehicle bottom monitoring scan image.

[0099] The method for stitching vehicle bottom images provided by this application, compared with the prior art, automatically matches the line scan camera's row frame P by using the vehicle speed matching method based on vehicle bottom monitoring v , which can enable the line scan camera to obtain continuous clear single-frame vehicle bottom images based on the optimal row frame required for image stitching, reducing the stitching difficulty and improving the quality of vehicle bottom image stitching.

[0100] Further, please refer to Figures 4 to 6 , the method for stitching the continuous single-frame images of the vehicle bottom includes the steps:

[0101] Detect the edge contour of the foreground target in the single-frame image;

[0102] Extract the center line of the edge contour of the foreground target;

[0103] Take the center line as a feature of the single-frame vehicle bottom image, and sequentially stitch the continuous single-frame vehicle bottom images.

[0104] It can be understood that, on the one hand, when the vehicle passes through the lane, the position of the vehicle is biased to one side, causing the foreground target (i.e., the local vehicle bottom) in the single-frame image not to achieve a centered effect. By extracting the center line of the edge contour of the foreground target, using the center line as a feature can play a role in centering and adjusting the foreground target.

[0105] On the other hand, when the vehicle passes through the lane and the steering wheel is not kept stable, there is jitter or turning. In this case, the existing splicing technology obtains a spliced image of the vehicle bottom, which is blurred and the vehicle bottom is bent and deformed. Therefore, in order to avoid the detection of the port law enforcement agency, when the vehicle passes through the inspection lane, some vehicles deliberately swing the steering wheel at a high frequency and cooperate with the vibration caused by the generator to avoid the bottom monitoring of the vehicle. In this embodiment, the single-frame images of the vehicle bottom are characterized by the center line, and the consecutive single-frame images of the vehicle bottom are spliced in sequence to ensure that the center line features in each single-frame image are on the same straight line after splicing, so as to correct the bent vehicle bottom image and improve the quality of the edge fitting effect, and a clear vehicle bottom scan image can be obtained even when the steering wheel of the vehicle is not kept stable.

[0106] Further, the method of sequentially splicing the consecutive single-frame images of the vehicle bottom further includes the steps of:

[0107] Cut the overlapping area of the single-frame image, find the stitching line with the minimum energy in the overlapping area and perform fusion processing.

[0108] It can be understood that, on the basis of taking the center line as the feature, a method of finding the best stitching line is also adopted, mainly to further solve the problems that cannot be solved in image alignment, such as large parallax and fusion ghost of moving objects in image splicing. By cutting the overlapping area of the images to be spliced, the stitching line with the minimum energy in the overlapping area can be found, and the spliced image is obtained by direct splicing.

[0109] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A vehicle speed matching method based on under-vehicle monitoring, characterized in that, Including: Step 1: Preset the line scan width x of the line scan camera and the pixel height y of a single frame. Step 2: Measure the actual maximum illumination field of view width X of the line scan camera, and calculate the actual length Y corresponding to a single-frame image. Step 3: Detect the speed V of the vehicle, and calculate the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image. Step 4: Automatically match the line-scan camera's line frame P according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image v ; In Step 2, the method for calculating the actual length Y corresponding to a single-frame image includes the steps of: Calculating the true distance s corresponding to each pixel, where: Based on the true distance s corresponding to each pixel, calculating the actual length Y corresponding to a single-frame image, where: Y = s·y; In Step 3, the method for detecting the speed V of the vehicle includes the steps of: Setting inductive loops A1 and A2 at the entrance of the lane; Pre-measuring the distance S between inductive loops A1 and A2; Sensing the time interval t between the vehicle triggering inductive loops A1 and A2; Calculating the speed V of the vehicle, where: In step four, automatically match the line-scan camera's row frame P according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image v , where:

2. The vehicle speed matching method based on underbody monitoring according to claim 1, wherein In Step 2, the method for measuring the actual maximum illumination field of view width X of the line scan camera includes the steps of: Installing the line scan camera after lens matching in the lane; Placing the calibration board parallel to the ground directly above the line scan camera; Moving the calibration board upward until the line scan camera obtains the maximum illumination field of view; Measuring the actual maximum illumination field of view width X corresponding to the maximum illumination field of view obtained by the scan camera on the calibration board.

3. The vehicle speed matching method based on underbody monitoring according to claim 1, wherein In Step 3, the speed V of the vehicle is detected in real time, and the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image is calculated in real time; In step four, the line scan camera's row frame P is automatically and real-time matched according to the time T required for the vehicle to pass through the actual length Y corresponding to a single-frame image v .

4. A method for stitching images of the vehicle bottom, characterized in that, Including the steps of: Automatically match the line scan camera's line frame P using the vehicle speed matching method based on under-vehicle monitoring according to any one of claims 1 to 3 v ; The line scan camera obtains continuous single-frame images of the vehicle bottom; Stitching the continuous single-frame images of the vehicle bottom; Outputting a complete monitoring scan image of the vehicle bottom.

5. The vehicle bottom image stitching method according to claim 4, wherein, When the line scan camera obtains continuous single-frame images of the vehicle bottom, the fill lights on both sides of the line scan camera are turned on simultaneously.

6. The vehicle bottom image stitching method according to claim 4, wherein The method for stitching the continuous single-frame images of the vehicle bottom includes the steps of: Detecting the edge contour of the foreground target in a single-frame image; Extracting the center line of the edge contour of the foreground target; Using the center line as a feature for the single-frame images of the vehicle bottom, and sequentially stitching the continuous single-frame images of the vehicle bottom.

7. The vehicle bottom image stitching method according to claim 6, wherein, The method for sequentially stitching the continuous single-frame images of the vehicle bottom further includes the steps of: Cutting the overlapping area of the single-frame images, finding the stitching line with the minimum energy in the overlapping area, and performing fusion processing.

Images