Adaptive Image Fusion Scheme Based on Contourlet Transform and Machine Learning

摘要

Adaptive image fusion scheme based on the combination of contourlet transform, Kernel Principal Component Analysis (K-PCA), Support Vector Machine (SVM) and Mutual Information (Ml) is proposed. Contourlet is well suited to image fusion scheme because of its properties, such as localization, midtiresolution, directionality and anisotropy. K-PCA operates on low frequency sub band to extract feature and SVM is applied to high frequency sub bands to obtain a composite image with extended information. Moreover, Mutual Information (MI) is used to adjust the contribution of each source image in the final fused image. Performance evaluation is carried out by using recently developed metric, Image Quality Index (IQI). The proposed scheme outperforms previous approaches both subjectively and quantitatively, and this is evident from the experimental results and findings.