Multi-focus Image Fusion Algorithm Based On Joint Bilateral Filter

In order to accurately analyze the information of image during the image processing,it is important to obtain images with all clear objects in focus.To solve the limitation of optical lens depth of field,multi-focus image fusion is used to combine two or more images with different focus areas to obtain a fully focused image with all objects in focus,which improves the utilization of the information contained in the source images and facilitates the subsequent image processing and analysis.It is widely used in digital photography,medical equipment,aerospace,digital image processing,and other fields.The joint bilateral filter,as an edge-preserving filter,can effectively retain the edge feature information of the guide image to the fused image.It plays a better role in edge-preserving smoothing.This paper proposes two multi-focus image fusion methods based on joint bilateral filter.(1)For the multi-focus image fusion algorithm based on focused region detection,the edges are easy to lose parts of the detail information,which makes the blurring and artifacts appear at the edge connection.A multi-focus image fusion method based on joint bilateral filter and random wander is proposed.The joint bilateral filter can effectively smooth the image detail information while preserving the image edge structures,improve the problem of artifacts and blurring at the edges of the fused image,and make the fused image more effective.Firstly,the source images are decomposed into large-scale and small-scale focus images by Gaussian filter,and the edges of the decomposed large-scale and small-scale focus images are smoothed by different guiding filters.Then,the large-scale and small-scale focus maps are used as the marker nodes of the random walk algorithm,the initial decision map is obtained by the fusion algorithm,and the guided filter is used to optimize the decision map again.Finally,the source images are reconstructed to obtain the final fused image according to the decision graphs.The experimental results show that this method better obtains the focused information in the source images,better preserves the edge texture and detail information in the focused region,and reduces the artifacts at the edge connection of the focused region.(2)To address the problem that some fusion methods tend to ignore the capture of luminance and contrast information,which makes the fusion result map appear weaken in luminance and contrast.A multi-focus image fusion method based on the multi-level decomposition of the fast joint bilateral filter is proposed.A new edge-preserving multilevel decomposition model is designed by combining the fast joint bilateral filter with Gaussian filter.Firstly,the source images are decomposed into three feature maps and one base layer map using the edge-preserving multilevel decomposition model.then,the feature maps are fused with the improved PCNN fusion rule and the region energy fusion rule,and the base layer is fused with the contrast saliency fusion rule.Finally,the fused image is obtained by overlaying and reconstructing each feature fusion map and the base layer fusion map.The experimental results show that the fused image of the proposed algorithm is better and can effectively improve the brightness and contrast of the fused image while preserving the edge structure well.This paper designs a multi-focus image fusion system based on the joint bilateral filter,and systematically analyzes and evaluates the performance of the above two methods from various aspects,and also selects a variety of mainstream multi-focus fusion methods to compare with this paper to verify the advantages of the proposed method,which also provides technical support for the wide application of multi-focus image fusion.