A Survey on Multivariate Methods for Multimodal Fusion of CT/MRI Images

Riya Gupta, Dr. C. S. Raghuvanshi

Medical CT and MR image fusion, adaptive structure decomposition (ASD), MI, SSIM, EN.

In many different application domains, image fusion is widely acknowledged as a helpful technique for enhancing overall system performance. These application fields include, but are not limited to, remote sensing, traffic control, machine learning, health care applications, combat surveillance, detection of disguised munitions, and identification of faults in non-destructive testing. The amount of scholarly literature on the topic of medical image fusion has significantly increased during the last several years. Image fusion has grown in importance as a component of the commonly used image processing applications because to the large variety of capture devices that are currently available. Image fusion is the process of matching relevant information from several sensors using various mathematical models to create a single composite image. Combining data from many sensors, multiple viewpoints, and various temporal dimensions into a single image is known as image fusion. This preserves the integrity of important characteristics while improving the quality of the image. Robot vision, aerial, satellite, and medical imaging, as well as robot and vehicle navigation, are just a few of the many applications that heavily depend on this stage. This study examines many cutting-edge picture fusion methods at various levels, along with the benefits and drawbacks of each. It also explores several transform-based and spatial approaches with quality measures and their applicability across various industries. Finally, this study's results have shown several potential future application avenues, such as picture fusion applications.