Image fusion based on adaptive enhancement and three-dimensional decomposition

To improve the detail structure, contrast, and overall visual quality of fused images, this paper introduces an infrared and visible image fusion method that leverages adaptive enhancement and three-scale decomposition, effectively addressing the insufficient use of source images found in current enhancement and fusion techniques. Initially, an enhanced CLAHE method is incorporated into improving the visible light image. Following this, the source images are split into base and structure layers through the L1-L0 method, and the base layer is further decomposed into base structure and base texture layers using the texture-structure method. The fusion of the base texture layers is performed using a method of taking the larger absolute value; the base structure layers are fused using an improved image moment-guided method; and in the detail layer, the information entropy weighted variance fusion method is adopted. Experimental results demonstrate that this method surpasses several existing fusion techniques in both subjective and objective evaluation metrics, higher AG/EN than baselines, the CC index is also quite close to the optimal value , and SD, SF, and VIF mostly rank among the top tier, significantly enhancing the detail clarity and visual quality of the fused images, showing the method's validity and practical applicability.