Research On Panoramic Image Mosaic Based On BRISK Image Registration

Image stitching technology in the field of image processing is of great significance,but also in the field of image processing research hot spots.In order to concatenate several images under the same scene into a panoramic image with a wide viewing angle,no obvious splicing gap,and a high degree of information,needs to apply the panorama image splicing technique.Panoramic image splicing has important applications in computer identification,medical image analysis and virtual reality technology.The main contents of this paper are divided into image registration and image fusion technology.On the basis of studying the existing algorithms,an improved algorithm is proposed for image registration and image fusion.This paper first describes the research background and significance of image stitching technology,and analyzes the current situation of research at home and abroad,introduces the basic process of image splicing,realizes BRISK image registration based on normalized product correlation,and based on color symbiosis multi-resolution image fusion algorithm for matrix panorama image stitching.For image registration,this paper presents a BRISK image registration algorithm based on normalized product correlation.The algorithm first uses the AGAST algorithm to extract the feature points,and then applies the normalized product correlation algorithm to block the image.The matching points are used to match the main points of the feature points and the descriptor vectors.Match the successful feature points,and use the random sample consistency method to remove the mismatched points.The simulation experiments of three sets of images with rotation and angle of view are carried out.The analysis of the images and data obtained by simulation shows that:In this paper,the improved BRISK image registration algorithm is compared with the original algorithm,the traditional SIFT algorithm and the advanced algorithm ORB,the feature point matching accuracy and matching speed are improved.After the image registration is completed,the geometric transformation relation between the images is calculated by the cylindrical projection algorithm,and then the matrix transformation between the image sequences is carried out to realize the splicing of multiple images.After the splicing is completed,the image fusion,in order to eliminate the splicing gap and ghosting and so on.For image fusion,this paper proposes a multi-resolution image fusion algorithm based on color co-occurrence matrix.The algorithm firstly performs the color co-occurrence matrix calculation on the image,extracts the texture features of the image overlapping region,and uses the extracted texture features to perform multi-resolution fusion.Image fusion experiment application fusion evaluation criteria analysis of the experimental results show that:The improved multi-resolution image fusion algorithm proposed in this paper has better fusion effect than the original algorithm,eliminating the splicing gap and ghosting,and realized the seamless splicing of the panoramic image.