CN104992406B - A kind of highway bridge image acquiring method of not close traffic - Google Patents
A kind of highway bridge image acquiring method of not close traffic Download PDFInfo
- Publication number
- CN104992406B CN104992406B CN201510332329.2A CN201510332329A CN104992406B CN 104992406 B CN104992406 B CN 104992406B CN 201510332329 A CN201510332329 A CN 201510332329A CN 104992406 B CN104992406 B CN 104992406B
- Authority
- CN
- China
- Prior art keywords
- bridge deck
- image
- pictures
- camera
- angle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Closed-Circuit Television Systems (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
本发明公开了一种不封闭交通的公路桥面图像获取方法，首先，使用定点摄像机按照设定的角度和焦距，来回扫描公路桥面获得图像；其次，通过对获得的图像进行识别和装拼，获得桥面各段不含车辆信息的图像；接着对各段桥面图像做畸变矫正，然后把各段桥面的图像重新拼接，最终形成完整的公路桥面图像。与传统的人工获取公路桥面破损信息方式相比，本发明提供了一种无封闭交通，非接触式快捷高效获取公路桥面病害信息方式，将为实现无需封闭交通而能对桥面进行质量检测，打下关键技术基础。与多功能路面信息采集车相比，本发明有着设备要求低且成本低廉的优势。
The invention discloses an image acquisition method of a highway bridge deck without closed traffic. First, a fixed-point camera is used to scan the highway bridge deck back and forth according to a set angle and focal length to obtain images; secondly, by identifying and assembling the acquired images, Obtain images of each section of the bridge deck without vehicle information; then perform distortion correction on each section of the bridge deck image, and then re-splicing the images of each section of the bridge deck to finally form a complete road bridge deck image. Compared with the traditional way of manually obtaining road bridge deck damage information, the present invention provides a non-closed traffic, non-contact fast and efficient way of acquiring road bridge deck disease information, which will be able to monitor the quality of the bridge deck without closed traffic. Testing, laying the foundation for key technologies. Compared with the multifunctional road surface information collection vehicle, the present invention has the advantages of low equipment requirements and low cost.
Description
技术领域technical field
本发明涉及公路桥面信息收集的技术领域，尤其是指一种不封闭交通的公路桥面图像获取方法。The invention relates to the technical field of road bridge deck information collection, in particular to a road bridge deck image acquisition method that does not close traffic.
背景技术Background technique
国家在公路桥梁的建设投入巨大，但是在维护公路桥梁的健康上也花费了大量经费。为数不少的桥梁在通车使用过程中，桥面不同程度地出现了裂缝、车辙、拥包、坑槽等病害，造成相当的经济损失，严重时还会发生交通事故。因此，对桥面质量的监测、评估和日常养护非常重要。The country has invested heavily in the construction of roads and bridges, but it has also spent a lot of money on maintaining the health of roads and bridges. During the opening to traffic of a large number of bridges, the bridge decks have cracks, ruts, pockets, potholes and other diseases to varying degrees, causing considerable economic losses, and even serious traffic accidents. Therefore, the monitoring, evaluation and daily maintenance of bridge deck quality are very important.
然而，目前国内各地对桥面铺装的监测与检测、质量评估和日常维护绝大部分仍然采用人工现场判断、定期检查和维护的方式，这种方式一是极其浪费人力物力，现场的勘测费时费工，效率低；二是需要封闭单个车道交通，甚至需要暂时完全封闭整个桥梁的通行，容易造成交通拥堵，因此，在不封闭桥梁交通的情况下，如何进行桥面质量的监测与评估是急需解决的问题。However, at present, most of the monitoring and testing, quality evaluation and daily maintenance of bridge deck pavement in various parts of the country still adopt the method of manual on-site judgment, regular inspection and maintenance. It is labor-intensive and inefficient; the second is that it is necessary to close a single lane of traffic, or even temporarily close the entire bridge, which is likely to cause traffic congestion. Therefore, how to monitor and evaluate the quality of the bridge deck without closing the bridge traffic is a Urgent problems.
这种工程需求驱动了非接触式测量方法的发展，对采集的桥面图像进行分析是非接触式测量方法中的一个重要组成部分。要对获得的图像进行病害信息研究，首先要解决如何清晰便捷获得路面图像问题。这是非接触式测量中能否成功实施的一个重要关键。目前已有多功能路面信息采集车，通过绑定在车尾的摄像机获取路面图像信息，但是由于造价昂贵，难以推广。This engineering requirement drives the development of non-contact measurement methods, and the analysis of collected bridge deck images is an important part of non-contact measurement methods. To conduct disease information research on the obtained images, the problem of how to obtain road surface images clearly and conveniently must be solved first. This is an important key to the successful implementation of non-contact measurements. At present, there are multifunctional road surface information collection vehicles, which can obtain road surface image information through a camera bound at the rear of the vehicle, but it is difficult to popularize due to the high cost.
发明内容Contents of the invention
本发明的目的在于克服现有技术的不足，针对当前路面破损探测技术存在检测成本高、效率低的局限性，提供一种不封闭交通的公路桥面图像获取方法，通过多次扫描获取公路桥面图像信息的技术，解决了如何在公路桥面病害信息检测中清晰便捷获得公路桥面信息的问题。The purpose of the present invention is to overcome the deficiencies of the prior art, aiming at the limitations of high detection cost and low efficiency in the current pavement damage detection technology, to provide a road bridge image acquisition method that does not close the traffic, and obtain the road bridge image through multiple scans. The technology of surface image information solves the problem of how to obtain highway bridge deck information clearly and conveniently in the detection of highway bridge deck disease information.
为实现上述目的，本发明所提供的技术方案为：一种不封闭交通的公路桥面图像获取方法，首先，使用定点摄像机按照设定的角度和焦距，来回扫描公路桥面获得图像；其次，通过对获得的图像进行识别和装拼，获得桥面各段不含车辆信息的图像；接着对各段桥面图像做畸变矫正，然后把各段桥面的图像重新拼接，最终形成完整的公路桥面图像。In order to achieve the above object, the technical solution provided by the present invention is: a method for acquiring images of road bridge decks without closed traffic. First, use a fixed-point camera to scan back and forth on the road bridge decks to obtain images according to the set angle and focal length; secondly, By identifying and assembling the obtained images, the images of each section of the bridge deck without vehicle information are obtained; then the distortion correction is performed on the images of each section of the bridge deck, and then the images of each section of the bridge deck are re-joined to form a complete highway bridge face image.
本发明所述不封闭交通的公路桥面图像获取方法，包括以下步骤：The highway bridge deck image acquisition method of non-closed traffic of the present invention, comprises the following steps:
1)事先统计摄像机拍摄覆盖范围，实际桥面的长度，得到搭架的摄像机数目，且在桥面高处搭建高分辨率摄像机前，得到摄像机旋转的角度及拍摄焦距参数；1) Calculate the camera shooting coverage area and the actual length of the bridge deck in advance to obtain the number of cameras erected, and before building a high-resolution camera at the height of the bridge deck, obtain the camera rotation angle and shooting focal length parameters;
2)每个摄像机从设定的起始角度开始拍摄，拍摄l张图片，旋转到下个角度，再拍摄l张图片，直至旋转到设定的结束角度；2) Each camera starts shooting from the set starting angle, takes l pictures, rotates to the next angle, and takes another l pictures until it rotates to the set end angle;
3)每个摄像机在每个角度拍摄得到的l张图片，从其中筛选出无车辆信息的图片，若多张图片无车辆信息，则任取一张，标记为合格；如果l张图片都存在车辆信息，则采用图像处理方法尝试从这l张图片中提取出无车辆信息的互不相同的区域装拼出一张新的完整桥段图片，若装拼得到的新图片能覆盖该摄像机在该角度所能拍摄到的完整桥面范围，则将新图片标记为合格，否则标记为不合格；3) From the l pictures taken by each camera at each angle, the pictures without vehicle information are screened out. If there are multiple pictures without vehicle information, one is selected and marked as qualified; if all the l pictures exist Vehicle information, use image processing methods to try to extract different areas without vehicle information from this l picture to assemble a new complete bridge section picture, if the new picture obtained by assembling can cover the camera in the If the complete range of the bridge deck can be photographed at this angle, the new picture will be marked as qualified, otherwise it will be marked as unqualified;
4)检查所有角度得到的图片是否都为合格，如果是，则跳到下面步骤5)，否则返回上面步骤3)；4) Check whether the pictures obtained from all angles are qualified, if yes, then skip to step 5) below, otherwise return to step 3) above;
5)从步骤4)中获得单个摄像机拍摄范围内的各段桥面的无车辆信息图片后，按照摄像机成像原理以及相关坐标转换关系，对不同位置得到的无车图片做畸变矫正；5) After obtaining the vehicle-free information pictures of each section of the bridge deck within the shooting range of a single camera in step 4), according to the camera imaging principle and the relevant coordinate conversion relationship, distortion correction is performed on the vehicle-free pictures obtained at different positions;
6)将步骤5)中得到的矫正后的图片做拼接处理，把相邻角度拍摄得到的图片拼接起来，形成覆盖单个摄像机的所有拍摄范围的公路桥面图片；6) The corrected pictures obtained in step 5) are spliced, and the pictures taken at adjacent angles are spliced together to form a highway bridge deck picture covering all shooting ranges of a single camera;
7)将所有摄像机得到的公路桥面图片装拼得到完整无车公路桥图片。7) Assemble the road bridge deck pictures obtained by all cameras to obtain a complete car-free road bridge picture.
在步骤5)中，矫正的具体算法如下：In step 5), the specific algorithm of correction is as follows:
5.1)定出源图像的四边形的四个顶点，给出目的图像四边形的四个顶点，然后求出源图像的四个顶点转为目的图像对应的四个顶点的转换矩阵，最后通过转换矩阵把源图像转为目的图像。使用OpenCV提供的函数cvGetPerspectiveTransform(CvPoint2D32fsrcTri,CvPoint2D32f dstTri,CvMat*warp_mat)得到转换矩阵，其中srcTri表示源图像四边形的四个顶点坐标，dstTri表示目的图像四边形四个顶点坐标，warp_mat表示指向一个3×3输出矩阵的指针；5.1) Determine the four vertices of the quadrilateral of the source image, give the four vertices of the quadrilateral of the destination image, then find the conversion matrix of the four vertices of the source image into the corresponding four vertices of the destination image, and finally transform the Convert the source image to the destination image. Use the function cvGetPerspectiveTransform(CvPoint2D32fsrcTri, CvPoint2D32f dstTri, CvMat*warp_mat) provided by OpenCV to get the transformation matrix, where srcTri represents the coordinates of the four vertices of the quadrilateral of the source image, dstTri represents the coordinates of the four vertices of the quadrilateral of the destination image, and warp_mat represents pointing to a 3×3 output a pointer to the matrix;
5.2)将得到的透视变换矩阵warp_mat作为参数传到cvWarpPerspective(constCvArr*src,CvArr*dst,const CvMat*map_matrix)中，得到转换后的目的图像，其中src为指向输入图像的指针，dst为指向目的图像的矩阵；5.2) Pass the obtained perspective transformation matrix warp_mat as a parameter to cvWarpPerspective(constCvArr*src, CvArr*dst, const CvMat*map_matrix) to obtain the converted destination image, where src is a pointer to the input image, and dst is the destination matrix of images;
5.3)将矫正后的图像按序号保留在原来的数组中，以便后续的图像拼接。5.3) Keep the rectified image in the original array according to the serial number, so as to facilitate the subsequent image splicing.
本发明与现有技术相比，具有如下优点与有益效果：Compared with the prior art, the present invention has the following advantages and beneficial effects:
1、与传统的人工获取公路桥面破损信息方式相比，本发明提供了一种无封闭交通，非接触式快捷高效获取公路桥面病害信息方式，将为实现无需封闭交通而能对桥面进行质量检测，打下关键技术基础。1. Compared with the traditional way of manually obtaining road bridge deck damage information, the present invention provides a non-closed traffic, non-contact fast and efficient way to obtain road bridge deck disease information, which will realize bridge deck damage without closed traffic. Carry out quality inspection and lay a key technical foundation.
2、与多功能路面信息采集车相比，本发明有着设备要求低且成本低廉的优势。2. Compared with the multifunctional road surface information collection vehicle, the present invention has the advantages of low equipment requirements and low cost.
附图说明Description of drawings
图1为安装在灯柱上的高清变焦扫描摄像机组模拟图。Figure 1 is a simulation diagram of a high-definition zoom scanning camera group installed on a lamp post.
图2为m台摄像机拍摄桥面的示意图。Figure 2 is a schematic diagram of m cameras capturing the bridge deck.
图3为单个摄像机在同一角度拍摄l张图的示意图。FIG. 3 is a schematic diagram of a single camera shooting one picture at the same angle.
图4为从单个摄像机在某一角度拍摄的l张图片中拼出无车桥段的示意图。Fig. 4 is a schematic diagram of spelling out a bridge section without vehicles from l pictures taken by a single camera at a certain angle.
图5为单个摄像机获取完整无车桥面的流程图。Figure 5 is a flow chart of acquiring a complete vehicle-free bridge deck with a single camera.
图6为m台摄像机获取完整公路桥面图像信息的流程图。Fig. 6 is a flow chart of acquiring complete image information of the highway bridge deck by m cameras.
图7a为畸变图像矫正前的示意图。Fig. 7a is a schematic diagram of a distorted image before correction.
图7b为畸变图像矫正后的示意图。Fig. 7b is a schematic diagram of the distorted image after correction.
图8为拍摄桥段的成像示意图。Figure 8 is a schematic diagram of the imaging of the bridge section.
图9a为一张桥面图片。Figure 9a is a picture of the bridge deck.
图9b为一张与图9a相邻的桥面图片。Figure 9b is a picture of the bridge deck adjacent to Figure 9a.
图9c为图9a与9b装拼得到的桥段图。Fig. 9c is a bridge segment diagram obtained by assembling Figs. 9a and 9b.
具体实施方式Detailed ways
下面结合具体实施例对本发明作进一步说明。The present invention will be further described below in conjunction with specific examples.
本实施例所述的不封闭交通的公路桥面图像获取方法，主要是将摄像机组安装在桥面高处，比如路灯的灯柱上等，通过旋转轴精确控制摄像机来回旋转，获取公路桥每个位置的信息，由于不封闭交通，所以需要多次扫描桥面，使用定点摄像机按照设定的角度和焦距，来回扫描公路桥面获得图像，对获得的图像作无车辆信息图片筛选或者装拼出无车辆信息图片，排除车辆干扰，获得桥面各段不含车辆信息的图像，对摄像机在不同转角拍摄得到的照片存在的透视畸变的问题，进行畸变矫正，最后各段桥面的图像重新拼接，形成一段完整的公路桥面图像。其包括以下步骤：The road bridge deck image acquisition method without closed traffic described in this embodiment is mainly to install the camera group on the high place of the bridge deck, such as on the lamp post of the street lamp, etc., and accurately control the camera to rotate back and forth through the rotation axis, so as to obtain every image of the road bridge. The information of each location, because the traffic is not closed, so it is necessary to scan the bridge deck multiple times, use the fixed-point camera to scan the highway bridge deck back and forth according to the set angle and focal length to obtain images, and filter or assemble the obtained images without vehicle information Produce pictures without vehicle information, eliminate vehicle interference, obtain images of each section of the bridge deck without vehicle information, correct the perspective distortion of the photos taken by the camera at different corners, and finally recreate the images of each section of the bridge deck splicing to form a complete road bridge image. It includes the following steps:
1)事先统计摄像机拍摄覆盖范围，实际桥面的长度，得到搭架的摄像机数目，且在桥面高处搭建高分辨率摄像机前，得到摄像机旋转的角度及拍摄焦距参数；1) Calculate the camera shooting coverage area and the actual length of the bridge deck in advance to obtain the number of cameras erected, and before building a high-resolution camera at the height of the bridge deck, obtain the camera rotation angle and shooting focal length parameters;
2)每个摄像机从设定的起始角度开始拍摄，拍摄l张图片，旋转到下个角度，再拍摄l张图片，直至旋转到设定的结束角度；2) Each camera starts shooting from the set starting angle, takes l pictures, rotates to the next angle, and takes another l pictures until it rotates to the set end angle;
3)每个摄像机在每个角度拍摄得到的l张图片，从其中筛选出无车辆信息的图片，若多张图片无车辆信息，则任取一张，标记为合格；如果l张图片都存在车辆信息，则采用图像处理方法尝试从这l张图片中提取出无车辆信息的互不相同的区域装拼出一张新的完整桥段图片，若装拼得到的新图片能覆盖该摄像机在该角度所能拍摄到的完整桥面范围，则将新图片标记为合格，否则标记为不合格；3) From the l pictures taken by each camera at each angle, the pictures without vehicle information are screened out. If there are multiple pictures without vehicle information, one is selected and marked as qualified; if all the l pictures exist Vehicle information, use image processing methods to try to extract different areas without vehicle information from this l picture to assemble a new complete bridge section picture, if the new picture obtained by assembling can cover the camera in the If the complete range of the bridge deck can be photographed at this angle, the new picture will be marked as qualified, otherwise it will be marked as unqualified;
4)检查所有角度得到的图片是否都为合格，如果是，则跳到下面步骤5)，否则返回上面步骤3)；4) Check whether the pictures obtained from all angles are qualified, if yes, then skip to step 5) below, otherwise return to step 3) above;
5)从步骤4)中获得单个摄像机拍摄范围内的各段桥面的无车辆信息图片后，按照摄像机成像原理以及相关坐标转换关系，对不同位置得到的无车图片做畸变矫正；5) After obtaining the vehicle-free information pictures of each section of the bridge deck within the shooting range of a single camera in step 4), according to the camera imaging principle and the relevant coordinate conversion relationship, distortion correction is performed on the vehicle-free pictures obtained at different positions;
6)将步骤5)中得到的矫正后的图片做拼接处理，把相邻角度拍摄得到的图片拼接起来，形成覆盖单个摄像机的所有拍摄范围的公路桥面图片；6) The corrected pictures obtained in step 5) are spliced, and the pictures taken at adjacent angles are spliced together to form a highway bridge deck picture covering all shooting ranges of a single camera;
7)将所有摄像机得到的公路桥面图片装拼得到完整无车公路桥图片。7) Assemble the road bridge deck pictures obtained by all cameras to obtain a complete car-free road bridge picture.
下面我们以具体实验为例，结合图1至图9对本发明上述不封闭交通的公路桥面图像获取方法进行具体说明，其情况如下：Below we take concrete experiment as example, in conjunction with Fig. 1 to Fig. 9, the highway bridge deck image acquisition method of the above-mentioned unclosed traffic of the present invention is described in detail, and its situation is as follows:
实验取景为一块沥青空地，其表面铺设材料与实际桥面相近。实验场景按36:1的比例缩小。实验中，以单个摄像机可以覆盖三车道为准作实验，单个车道实际宽度取3.6m,实验中取0.1m；车道间分界线实际取0.15m，实验中取0.004m；灯柱实际高度取10m,实验中取0.28m；路灯间距取40m,实验中取1.11m；车辆模型是实际车辆大小的1/36；实验中使用两台摄像机，实验中桥长度为2.22m；图1是实际中摄像搭设模拟图；实验的拍摄示意图如图2所示。以下是实施步骤：The scene of the experiment is an asphalt vacant land, and its surface paving material is similar to the actual bridge deck. The experimental scene is scaled down by a ratio of 36:1. In the experiment, a single camera can cover three lanes for the experiment. The actual width of a single lane is 3.6m, and 0.1m in the experiment; the boundary between lanes is actually 0.15m, and 0.004m in the experiment; the actual height of the lamp post is 10m , take 0.28m in the experiment; the distance between street lights is 40m, and 1.11m in the experiment; the vehicle model is 1/36 of the actual vehicle size; two cameras are used in the experiment, and the bridge length in the experiment is 2.22m; Set up the simulation map; the shooting schematic diagram of the experiment is shown in Figure 2. Here are the steps to implement:
一、搭设实验场景1. Set up the experimental scene
使用双面胶带做车道分界线和公路边界，模拟三车道公路；在公路两边以三角架为模拟灯柱，调节到合适高度，固定摄像机于三角架上的云台。通过云台精确旋转摄像机。Use double-sided tape to make lane boundaries and road boundaries to simulate a three-lane road; use a tripod as a simulated lamp post on both sides of the road, adjust to a suitable height, and fix the camera on the tripod head. Precise camera rotation via gimbal.
二、进行实验2. Conduct experiments
每个摄像机的拍摄流程如图5所示，而m台摄像机获取完整公路桥面图像信息的流程如图6所示。其中，获取覆盖单个摄像机完整拍摄范围的无车辆信息的桥面图片的具体步骤如下：The shooting process of each camera is shown in Figure 5, and the process of obtaining the complete road bridge image information by m cameras is shown in Figure 6. Among them, the specific steps of obtaining bridge deck pictures without vehicle information covering the complete shooting range of a single camera are as follows:
A、每个摄像机在每个角度拍摄得到的l(实验中l取3)张图片，如图3所示，分别识别是否有车辆存在，如果其中一张图片不含车辆信息，如图4所示，标记此角度获得的图片合格；如果l张图片都存在车辆信息，则从这l张图片中提取出无车辆信息的并且互不相同的区域装拼出一张新的完整桥段图片，如果装拼得到的新图片能覆盖该摄像机在该角度所能拍摄到的完整桥面范围，则将新图片标记合格，否则标记为不合格。A. Each camera captures l (in the experiment, l takes 3) pictures obtained at each angle, as shown in Figure 3, respectively identify whether there is a vehicle, if one of the pictures does not contain vehicle information, as shown in Figure 4 Indicates that the pictures obtained by marking this angle are qualified; if there is vehicle information in the l pictures, then extract the non-vehicle information and different areas from the l pictures to assemble a new complete bridge section picture, If the assembled new picture can cover the complete range of the bridge deck that the camera can capture at this angle, the new picture is marked as qualified, otherwise it is marked as unqualified.
B、检查单个摄像机从始端到末端拍摄得到的n(实验中取5)张图片是否都标记为合格，如果是，停止拍摄，转下面步骤C；如果不都为合格，摄像机重新旋转到始端，开始新一轮拍摄。B. Check whether the n (take 5 in the experiment) pictures obtained by a single camera from the beginning to the end are all marked as qualified, if yes, stop shooting, and turn to step C below; if not all qualified, the camera rotates to the beginning again, Start a new round of shooting.
C、将得到无车辆信息图片数组传入后台，作畸变矫正和拼接处理。C. Transfer the obtained image array without vehicle information to the background for distortion correction and splicing.
由成像原理可知，摄像机拍摄的桥面会成近大远小的实像，如图7a所示，畸变矫正的目的是将畸变的图像矫正为图7b所示的图像，即得到桥面的俯拍图。图8是公路桥路段拍摄光路示意图，其中四边形ABCD是摄像机覆盖路面的实际范围，A1B1C1D1为ABCD所成像。From the principle of imaging, we can see that the bridge deck captured by the camera will become a real image with a large distance and a large distance, as shown in Figure 7a. The purpose of distortion correction is to correct the distorted image into the image shown in Figure 7b, that is, to obtain an overhead shot of the bridge deck. picture. Fig. 8 is a schematic diagram of the optical path for shooting on the highway bridge section, where the quadrilateral ABCD is the actual range covered by the camera on the road surface, and A 1 B 1 C 1 D 1 is the image formed by ABCD.
矫正的具体算法如下：The specific algorithm of correction is as follows:
A、定出源图像的四边形的四个顶点，如图8中的A、B、C、D四个顶点，给出目的图像四边形的四个顶点，然后求出源图像的四个顶点转为目的图像对应的四个顶点的转换矩阵，最后通过转换矩阵把源图像转为目的图像。使用OpenCV提供的函数cvGetPerspectiveTransform(CvPoint2D32f srcTri,CvPoint2D32f dstTri,CvMat*warp_mat)得到转换矩阵，其中srcTri表示源图像四边形的四个顶点坐标，dstTri表示目的图像四边形四个顶点坐标，warp_mat表示指向一个3×3输出矩阵的指针。A, determine four vertices of the quadrilateral of the source image, as shown in Fig. 8 four vertices of A, B, C, D, provide four vertices of the quadrilateral of the purpose image, then find four vertices of the source image and convert to The transformation matrix of the four vertices corresponding to the destination image, and finally convert the source image to the destination image through the transformation matrix. Use the function cvGetPerspectiveTransform(CvPoint2D32f srcTri, CvPoint2D32f dstTri, CvMat*warp_mat) provided by OpenCV to get the transformation matrix, where srcTri represents the coordinates of the four vertices of the quadrilateral of the source image, dstTri represents the coordinates of the four vertices of the quadrilateral of the destination image, and warp_mat represents pointing to a 3×3 Pointer to the output matrix.
B、将得到的透视变换矩阵warp_mat作为参数传到cvWarpPerspective(constCvArr*src,CvArr*dst,const CvMat*map_matrix)中，得到转换后的目的图像。其中src为指向输入图像的指针，dst为指向目的图像的矩阵。B. Pass the obtained perspective transformation matrix warp_mat as a parameter to cvWarpPerspective(constCvArr*src, CvArr*dst, const CvMat*map_matrix) to obtain the converted target image. Where src is a pointer to the input image, and dst is a matrix pointing to the destination image.
C、将矫正后的图像按序号保留在原来的数组中，以便后续的图像拼接。C. Keep the rectified image in the original array according to the serial number, so as to facilitate the subsequent image splicing.
对得到的单个摄像机在各个拍摄角度的无车辆信息桥面图片进行畸变矫正后，将拍摄角度相邻的图片拼接，如图9a至9c所示，其中图9a、9b是相邻的桥面图片，图9c是图9a、9b装拼得到的桥段图。将单个摄像机拍摄得到桥段图片依次装拼，从而得到覆盖单个摄像机完整拍摄范围的桥段图。将相邻摄像机得到的桥段图片传入后台，拼接在一起，形成完整的桥面图片。After distortion correction is performed on the bridge deck pictures without vehicle information obtained by a single camera at various shooting angles, the pictures with adjacent shooting angles are stitched together, as shown in Figures 9a to 9c, where Figures 9a and 9b are adjacent bridge deck pictures , Fig. 9c is a bridge section diagram obtained by assembling Fig. 9a and 9b. The pictures of bridge sections captured by a single camera are assembled sequentially to obtain a bridge section map covering the entire shooting range of a single camera. Pass the pictures of the bridge section obtained by the adjacent cameras to the background and stitch them together to form a complete picture of the bridge deck.
以上所述之实施例子只为本发明之较佳实施例，并非以此限制本发明的实施范围，故凡依本发明之形状、原理所作的变化，均应涵盖在本发明的保护范围内。The implementation examples described above are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Therefore, all changes made according to the shape and principle of the present invention should be covered within the scope of protection of the present invention.
Claims (2)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510332329.2A CN104992406B (en) | 2015-06-16 | 2015-06-16 | A kind of highway bridge image acquiring method of not close traffic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510332329.2A CN104992406B (en) | 2015-06-16 | 2015-06-16 | A kind of highway bridge image acquiring method of not close traffic |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104992406A CN104992406A (en) | 2015-10-21 |
| CN104992406B true CN104992406B (en) | 2018-04-13 |
Family
ID=54304214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510332329.2A Active CN104992406B (en) | 2015-06-16 | 2015-06-16 | A kind of highway bridge image acquiring method of not close traffic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104992406B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109614959B (en) * | 2019-01-09 | 2022-10-21 | 高顺河 | Highway bridge deck image acquisition method |
| CN110084115A (en) * | 2019-03-22 | 2019-08-02 | 江苏现代工程检测有限公司 | Pavement detection method based on multidimensional information probabilistic model |
| CN110969576B (en) * | 2019-11-13 | 2021-09-03 | 同济大学 | Highway pavement image splicing method based on roadside PTZ camera |
| CN111915910A (en) * | 2020-08-14 | 2020-11-10 | 山东领军智能交通科技有限公司 | Road traffic signal lamp based on Internet of things |
| CN112220448B (en) * | 2020-10-14 | 2022-04-22 | 北京鹰瞳科技发展股份有限公司 | Fundus camera and fundus image synthesis method |
| CN112885096A (en) * | 2021-02-05 | 2021-06-01 | 同济大学 | Bridge floor traffic flow full-view-field sensing system and method depending on bridge arch ribs |
| CN114359054B (en) * | 2022-01-07 | 2025-03-14 | 重庆交通大学 | A method for constructing bridge structure contour based on fixed-axis rotation photography |
| CN115901790A (en) * | 2022-12-29 | 2023-04-04 | 中国土木工程集团有限公司 | Bridge bottom defect detection system |
| CN118014907B (en) * | 2024-02-05 | 2025-06-20 | 招商局重庆交通科研设计院有限公司 | Stitching method based on image correction and shooting path |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593678A (en) * | 2013-10-16 | 2014-02-19 | 长安大学 | Long-span bridge vehicle dynamic load distribution detection method |
| CN104599249A (en) * | 2015-01-13 | 2015-05-06 | 重庆大学 | Cable rope bearing bridge deck vehicle load distribution real-time detection method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9014415B2 (en) * | 2010-04-22 | 2015-04-21 | The University Of North Carolina At Charlotte | Spatially integrated aerial photography for bridge, structure, and environmental monitoring |
| US9036861B2 (en) * | 2010-04-22 | 2015-05-19 | The University Of North Carolina At Charlotte | Method and system for remotely inspecting bridges and other structures |
| US20140336928A1 (en) * | 2013-05-10 | 2014-11-13 | Michael L. Scott | System and Method of Automated Civil Infrastructure Metrology for Inspection, Analysis, and Information Modeling |
-
2015
- 2015-06-16 CN CN201510332329.2A patent/CN104992406B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593678A (en) * | 2013-10-16 | 2014-02-19 | 长安大学 | Long-span bridge vehicle dynamic load distribution detection method |
| CN104599249A (en) * | 2015-01-13 | 2015-05-06 | 重庆大学 | Cable rope bearing bridge deck vehicle load distribution real-time detection method |
Non-Patent Citations (4)
| Title |
|---|
| Data analysis and visualization for the bridge deck inspection and evaluation robotic system;Hung Manh La等;《Visualization in Engineering》;20150224;第3卷(第6期);第1-16页正文第2页第2栏倒数第2-3段，第8页第1栏第2-3段，图2-图3，图6 * |
| Multi-Image Stitching and Scene Reconstruction for Evaluating Defect Evolution in Structures;Mohammad R. Jahanshahi等;《Structural Health Monitoring》;20110131;第10卷(第6期);第1-27页摘要、正文第3页倒数第1段，图1 * |
| 基于计算机视觉的桥梁表面缺陷检测技术研究;曾燕华;《中国优秀博硕士学位论文全文数据库（硕士） 工程科技Ⅱ辑》;20051015(第06期);第C034-118页 * |
| 智能化桥检车图像采集分析系统设计与研究;李淑慧;《现代电子技术》;20100201(第3期);第154-156+160页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104992406A (en) | 2015-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104992406B (en) | A kind of highway bridge image acquiring method of not close traffic | |
| CN110197203B (en) | Classification and identification method of bridge pavement cracks based on width learning neural network | |
| CN109300126B (en) | High-precision intelligent detection method for bridge diseases based on spatial positions | |
| CN105956942B (en) | Quality detection method and detection device for steel mesh of power pipe gallery based on machine vision | |
| CN105516654A (en) | Scene-structure-analysis-based urban monitoring video fusion method | |
| CN103714343A (en) | Method for splicing and homogenizing road face images collected by double-linear-array cameras under linear laser illumination condition | |
| CN209181784U (en) | A video measuring device applied to automatic parking system | |
| CN110390256B (en) | Asphalt pavement crack extraction method | |
| CN106677037B (en) | Portable Asphalt Pavement Disease Detection Method and Device Based on Machine Vision | |
| CN110444043B (en) | Parking space inspection system and method based on positioning technology | |
| CN114627107B (en) | Full-morphology detection method and system for bridge beam bottom diseases | |
| CN106781680A (en) | A kind of curb parking intelligent control method based on the detection of image empty parking space | |
| CN102609924A (en) | Digital high-fidelity image collecting and processing method of stone cave temple wall paintings | |
| CN110363706A (en) | A large-area bridge deck image stitching method | |
| CN117499585A (en) | A precise positioning maintenance inspection system and method based on road network monitoring | |
| CN118818447A (en) | A joint calibration method based on roadside radar and camera | |
| CN110969576B (en) | Highway pavement image splicing method based on roadside PTZ camera | |
| CN105910585A (en) | Rapid inspection and measuring method of illegal buildings based on oblique photography | |
| CN117760388A (en) | Bridge BIM measurement and construction planning method integrating laser scanning and drone oblique photography | |
| Yu et al. | Nuclear containment damage detection and visualization positioning based on YOLOv5m-FFC | |
| CN102779341B (en) | Novel method for identifying support position in pit construction process | |
| CN114529808A (en) | Pipeline detection panoramic shooting processing method | |
| CN117710843A (en) | Intersection dynamic signal timing scheme detection method based on unmanned aerial vehicle video | |
| CN114359652B (en) | A positioning detection method based on deep learning and monocular large field of view plane measurement | |
| CN116429023A (en) | Method and system for measuring three-dimensional pavement data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |