Research On The Quantitative Expression Of The Field Of View Relationship Under Close-range Photogrammetry

When the camera calibration method is used to describe the relationship between different field of view,the calculation results are often inaccurate and the real-time performance is poor because of the fixed form,poor flexibility,lack of robustness,and high requirements of the test platform.Therefore,from the perspective of image processing,combined with change detection and image matching technology,this paper proposes a convenient quantitative description method of large and small field of view by using change detection and image matching technology,which provides technical support for target recognition and analysis in a certain spatial range.The premise of solving the field-of-view relationship is to detect t he changes in the field of view.When the change area of the field of view is detected in the large field of view,a frame of original photograph is obtained from the camera in the large field of view.The camera i n the small field of view makes regular rotation and translation motion to take a series of sample photographs as a collection of sample photographs.This paper chooses VIBE algorithm as the change detection method,and uses the similarity measure of foreground and neighborhood background histogram to detect and eliminate “ghost” phenomenon in VIBE algorithm.In this paper,we use the photo matching technology to obtain the sample photo with the highest matching degree with the original photo and the coordinates of the homonymous image points between them.In this regard,the paper focuses on the method of matching based on image feature points.In terms of performance and detection effect,SURF algorithm is selected for photo matching.At the same time,the foreground image is used to approximate the original image for matching.On the basis of satisfying the matching accuracy,the matching time is about one fifth of the time consumed by the original photo matching,which greatly improves the matching efficiency.In order to restore the camera's space pose in the small field of view is relative to the space pose of the camera in the large field of view at Instant of the photos being photographed.The paper uses rlative orientation with the sample photos and original photos to obtain the pose parameters of the photo taken in the small field of view relative to the photo in the large field of view.At the same time,this parameter is used to correct the pose of the camera in the small field of view,so that the camera in the large field of view can align with the changing area in the field of view at the same time so as to work together to obtain high-quality video streams.Through the analysis of the video stream to meet the security requirements in the field of view,the comparative experiments show that the proposed algorithm has better robustness and feasibility,and has certain guiding significance for realizing the security of the field of view.