Research On Large Field Of View And High Resolution Image Mosaic And Target Tracking Technology

With the rapid development of social economy and science and technology,more and more places and places have begun to install and use various types of video surveillance equipment.However,most of the existing video surveillance devices are captured by shooting at a fixed location,and the video images acquired by a single camera device have a limited field of view,and there are visual blind areas,which may pose potential safety hazards.With the application of image splicing technology,multiple cameras can now be combined to form a video surveillance system with a large field of view and no dead ends.At the same time,in the large field of view obtained by splicing,it is often necessary to detect and track specific targets,but there are few related studies in this area.Therefore,this paper studies the large-field-of-view high-resolution image stitching and target tracking technology.In view of the slow processing speed and low efficiency of traditional image stitching methods,which cannot meet the needs of fast and accurate stitching of high-resolution images,this paper proposes an improved algorithm for high-resolution image stitching based on ORB features.First,on the basis of ORB feature point extraction,Hamming distance is used for fast rough matching,and then the matching point pairs are optimized by the progressive sampling consistency(PROSAC)algorithm,after removing the mismatched point pairs,the image transformation matrix is solved.Finally,the weighted fusion algorithm is used to fuse the overlapping areas of the image,and the stitching traces are removed.The experimental results show that compared with the traditional ORB algorithm,the matching time of the feature points of the algorithm in this paper is only 65.10%,the matching accuracy rate is increased by 4.04%,the root mean square error is reduced by20.28%,and the image entropy is increased by 1.60%.The algorithm in this paper not only has obvious advantages in processing speed,but also has higher matching accuracy,which can realize fast and accurate stitching of high-resolution images.Through research and design,a large field of view and high-resolution image splicing system is designed,and its composed of camera imaging unit and main control unit are described in detail.The camera imaging unit is mainly composed of multiple visible light cameras.The selection of the imaging chip and the design of the chip drive are described in detail.Multiple cameras are arranged according to a special angle to collect image data to achieve high-resolution imaging detection with a large field of view.The main control unit is mainly composed of an image signal acquisition module,an image stitching module and a target tracking module.Firstly,it completes the acquisition and synchronization of the camera image signal,and then performs image stitching on the images collected by the camera array.Complete the tracking of the target.The system experiment platform was built,and two parts of experiments were conducted to verify,namely,large field of view and high resolution image stitching experiment and video image stitching and target tracking.The experimental results show that the system can quickly splice large field of view and high resolution video images.The field of view is large,the resolution is high,and the frame rate can be maintained at about30 frames.The splicing effect is better,the system is stable,the real-time performance is high,and the overall performance is better.The video image splicing and target tracking system can maintain a high frame rate,and can run stably for a long time,which can meet the needs of actual conditions.