Real-Time Image Fusion Using Nature-Inspired Optimization Techniques

Abstract

This paper presents a real-time image fusion method utilizing a nature-inspired optimization technique. The approach enhances adaptability by employing an objective function to determine the most suitable parameters for image fusion processes. To accomplish this, various techniques such as bilateral and speckle filtering, gamma correction for image enhancement, and fusion through stationary wavelet transform (SWT) are integrated. The optimization of these techniques is guided by the Yellow Saddle Goatfish (YSG) algorithm, which mimics the foraging behavior of the fish. The YSG algorithm stands out among other nature-inspired methods due to its strong exploration capabilities, making it effective in identifying optimal solutions. The proposed method is assessed using both quantitative metrics and visual evaluation. The findings reveal that the optimal parameters differ for each image, indicating the YSG algorithm’s ability to adapt based on image-specific characteristics. Moreover, the experimental results confirm that the proposed fusion model produces higher-quality fused images compared to traditional approaches.