Research On Camera Calibration Method In Video Detection

With the rapid development of modern industry, the traditional tools and methods ofdetection have become increasingly unable to meet the requirements of measurement, theprecision measurement technology needs to be developed eagerly. Visual measurement asa new measurement idea, is able to adapt to a variety of poor measurement environment,the precise positioning of a space point also ensures adequate measurement accuracy, itshigh-speed development meets the new requirements of modern industrial production anddetection in the measurement accuracy and measurement range.The main task of the vision measurement system is to use the computer to describeand identify three-dimensional objects. To accomplish these tasks, the visual measurementsystem needs to establish a relational model between the three-dimensional scene pointsand camera image pixel. The process that solving the parameters in this relational modelwith the world coordinates and image pixel coordinates of the known feature points, isknown as the camera calibration. The parameters of the relational model include cameraextrinsic parameters that describe the camera position and attitude information and cameraintrinsic parameters that describe the camera main point, the grid interconnection distanceand focal length.This paper mainly studies camera calibration issues of video detection inengineering applications and according to the characteristics and requirements of videoinspection of nuclear power plants, we presents a simple and convenient cameracalibration method. Firstly, according to the relationship between the camera coordinatesystem and world coordinate system when the camera adjusts its poses, we haveestablished the camera imaging model. Secondly, in response to the phenomenon thatcamera may deflects around the camera optical axis and brings error into the cameraimaging model, we have corrected the pixel coordinates of the feature points, and improve the method. Finally, we have dealt with the pixel coordinates of feature points withoptimal solution method, and solve the problem of Gaussian noise, and then get theaccurate, stable camera parameters. Apply this method to camera calibration, which isused in nuclear power plant, the camera intrinsic parameters applied by this methodimprove the measurement accuracy of a video measuring system to detect defects.