Research on Image Fusion Technology Based on Compressed Sensing

With the advancement of compressed sensing technology, research on compressed sensing-based image fusion has gained increasing attention. Leveraging the characteristics of image Fourier transform coefficients, this study proposes a compressed sensing domain image fusion algorithm based on high-frequency and low-frequency importance metrics under a dual-star sampling mode. The algorithm begins by acquiring measurements through dual-star sampling, then calculates importance metrics for high- and low-frequency regions as fusion operators, performs weighted fusion of the measurements, and finally reconstructs the fused image by solving a minimum total variation optimization problem. Subjective and objective experimental results demonstrate that this algorithm outperforms other Fourier-based approaches, with implementations involving sparse sampling and convex optimization techniques.