A Curvelet-Transform-Based Image Fusion Method Incorporating Side-Scan Sonar Image Features

Abstract

Current methods of fusing side-scan sonar images fail to tackle the issues of shadow removal, preservation of information from adjacent strip images, and maintenance of image clarity and contrast. To address these deficiencies, a novel curvelet-transform-based approach that integrates the complementary attribute of details from side-scan sonar strip images is proposed. By capitalizing on the multiple scales and orientations of the curvelet transform and its intricate hierarchical nature, myriad fusion rules were applied at the corresponding frequency levels, enabling a more-tailored image fusion technique for side-scan sonar imagery. The experimental results validated the effectiveness of this method in preserving valuable information from side-scan sonar images, reducing the presence of shadows and ensuring both clarity and contrast in the fused images. By meeting the aforementioned challenges encountered in existing methodologies, this approach demonstrated great practical significance.

Semantic-Aware Fusion Network Based on Super-Resolution

“…Among them, the multi-scale transform-based fusion framework is mainly a method to decompose the source image into different scales of information and analyze and fuse them at these scales. Commonly used multi-scale transform methods include wavelet transforms [ 6 , 7 , 8 ] (e.g., discrete wavelets), pyramid transforms [ 9 , 10 , 11 , 12 ] (e.g., gaussian pyramid or laplace pyramid), and methods based on multi-scale geometric analysis [ 13 , 14 , 15 ] (e.g., contour transform or curve transform). These transform methods are able to capture both detailed and global information in an image and have different applicability to different types of images.…”

“…There are more current infrared and visible image fusion methods, but they are mainly categorized into two groups: traditional methods and deep learning (DL)-based methods. Traditional fusion methods are usually based on fusion in the spatial and transform domains [ 5 ], and the image fusion frameworks used mainly include multi-scale transform (MST)-based fusion frameworks [ 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ], sparse representation (SR)-based fusion frameworks [ 16 , 17 , 18 ], subspace-based fusion frameworks [ 19 , 20 , 21 ], saliency-based fusion frameworks [ 22 ], and hybrid fusion frameworks [ 23 , 24 , 25 ]. And according to the adopted network architecture, the DL-based image fusion methods can be mainly categorized into three groups, which are autoencoders (AE)-based image fusion frameworks [ 26 , 27 , 28 , 29 , 30 ], convolutional neural network (CNN)-based image fusion frameworks [ 31 , 32 , 33 , 34 , 35 , 36 , 37 ] and generative adversarial network-(GAN) based image fusion frameworks [ 38 , 39 , 40 , 41 , 42 , 43 ].…”