Target Recognition And Location Based On Monocular Vision

In recent years,advances in computer and image processing technology have enabled the use of computer vision to simulate biomedical vision.The visual technology has developed rapidly and has been widely used in industrial,medical,aerospace,robot navigation and target tracking Has been widely used.The key problem of visual technology research is the recognition and location of the target.Accurate recognition and localization are the important indexes to measure the visual system.Monocular vision is an important branch of visual technology.It has the advantages of simple structure and flexible operation,so it has become a research hotspot of visual technology.In this paper,the research and development of vision technology are summarized and analyzed,and the key issues of monocular vision systems,such as target recognition,camera calibration and target location,are studied emphatically.Visual target recognition and location method,has a certain practical value.In this paper,we focus on monocular vision based target recognition and location technology,and accomplish the following tasks:Firstly,this paper studies and summarizes the target recognition technology based on image feature.The classical Harris corner detection algorithm and SIFT algorithm are introduced in detail.And in view of SIFT algorithm's feature matching accuracy is not high,the matching time is too long In this paper,an improved algorithm is proposed.Based on the idea of feature classification,this algorithm improves the traditional SIFT algorithm by using the corner feature,reduces the number of feature points,and improves the real-time performance of the algorithm.Compared with the traditional SIFT algorithm,the extracted feature points have higher significance,and the matching speed and accuracy are improved.Secondly,the camera calibration is the basis of visual technology research.In this paper,the basic principles of camera imaging,the linear and nonlinear models of the camera are introduced in detail.The typical camera calibration methods are analyzed and summarized,and the camera calibration is carried out based on the classic Zhang camera two-step template method.The internal parameters and the distortion coefficient of the camera are calculated and the camera is calibrated accurately.Thirdly,the problem of target location based on monocular vision is studied.Based on the analysis and comparison of the monocular vision localization method,two image-based methods are used to locate the target.The method obtains the target scene image from different viewpoints by changing the position of the camera,and obtains the basic matrix and the essence matrix by the relationship between the geometric constraint of the polar line and the matching points between the images,and obtains the relative positional relationship between the two cameras.target setting.The experimental results show that the proposed algorithm has a good accuracy of target location in a small scale,and can meet the specific requirements and has certain practical value.