Research On The Video-based Detection System For Tunnel Luminance

With the constantly advancing of transportation network coverage throughout the country, the number of highway tunnels increases and construction scale expands gradually. The role played by tunnels is becoming more critical. The detection of luminance and luminance uniformity for tunnels has important implications for reducing energy consumption and ensuring the safety traveling. Therefore, it’s necessary to develop a luminance measurement system to make up the shortcomings of traditional detectors.The thesis presents a video-based tunnel luminance detection system based on prevailing tunnel video surveillance system, and supported by video detection technology. Firstly, the system architecture and essential algorithms of software design are determined on the basis of principle of image acquisition and characteristics of video images. The essential algorithms are divided into two categories, video image stability and uniformity and luminance detection. Secondly, aiming at video image stability and uniformity algorithm, this thesis proposes video image stability detection algorithm, to ensure the camera has a good stability. Then uses the checkerboard calibration method to measure the distortion of the video taken by camera, meanwhile detects and corrects the uniformity of the camera image sensor, completing the pre-processed of video image. Next, luminance detection algorithms are studied, this thesis chooses median filtering method to extract the initial background image, ensuring the detection area accurately determined. The method of constant interval of two lines and two columns is selected to compute the image element. The algorithm model of luminance detection is determined by analyzing the results of detected image element and actual luminance. Finally, the algorithms are implemented by using VS2010,MFC library, OpenCV library and TeeChart Pro ActiveX controls, and then achieve the detection of luminance.Test results show that the system can detect the luminance and luminance uniformity, by processing the video images. The results also demonstrate that the precision of luminance detection system meet the testing requirements.