Fractional wavelet combined with multi-scale morphology and PCNN hybrid algorithm for grayscale image fusion

Grayscale image fusion is an important part of the digital image processing field, which is important for the integration of image information. This paper proposes a hybrid algorithm that addresses the problem of unclear edges caused by traditional image fusion algorithms. The proposed algorithm combines the discrete fractional wavelet transform with multi-scale morphology and the pulse coupled neural network. The hybrid algorithm employs discrete fractional wavelet transform to decompose the source images and obtain subbands that include both high- and low-frequency components. An image enhancement method, enhanced by multi-scale morphological operations, is developed to process the low-frequency subband. Additionally, a simplified pulse coupled neural network method is employed to adapt the high-frequency components and generate the high-frequency decision map. Fused images show that proposed algorithm effectively suppresses the Gibbs effect. Simulation experiments confirm that the fusion effect of the hybrid algorithm in this paper is better than the existing five classical algorithms, indicating that the hybrid algorithm is an efficient grayscale image fusion method.