Image Measuring Technology For Position Parameter Of Low-altitude Flying Object

In this paper, image acquiring technology for position parameter of low-altitude flying object is researched for the most part, seeing that flying object parameters needs to be recorded all the time in the spacecraft launching site. The principle is that solve object position parameter and off-course amount according to the images analysis obtained from different viewpoint. Binocular stereo vision measurement is a new measuring method based on optical measuring principles. Nowadays as the rapid development of computer processing and image acquiring technique, it makes this technique possible from algorithm and software facets. Stereo measurement technology based on computer vision has been widely applied in many fields, such as aviation mapping, military application, mechanical manufacture, biology technique and so on. And the Stereo measurement has become an important part in the measurement field.Principle of optical imaging system, coordinates transform relations and the main image distortion are elaborated first of all. A rectifying method was advanced after analyzing the perspective distortion caused by the camera angle. And it solved the measuring deviation problem caused by camera angle in low-altitude object measurement. Research method to pose camera for that the image measuring precision is affected by corresponding situation relationship between two cameras.After analyzing the traditional camera calibration methods, research the calibration method, rectify the perspective distortion and then calibrate the camera using two-step method for perspective distortion and nonlinear distortion exist in imaging system.For characteristics extraction problem, Moravec algorithm, Harris algorithm, SUSAN algorithm and Forstner algorithm were researched and compared and choose Harris for corner detection.Advanced the principles and methods, and then image measuring system is accomplished. And then the orientation algorithm is accomplished in Visual C++, and the experiment testified that the measuring system and methods are valid.