Otsu Thresholding based Image Fusion Framework using Contour-let Transform

Nowadays the result of infrared (I-R) and visible (V-I) image fusion is significantly used in major applications such as surveillance, remote sensing, military, etc. in order to enhance visibility and improve situation awareness. Most importantly fusion algorithm is applied on those I-R image and V-I image that capture the same scene information during low light and adverse weather conditions such as fog, snow, and dust. This paper is presenting a novel image-fusion framework for improving scene information that is captured during low light and adverse weather conditions. In proposed algorithm the contrast of the source images is improved by using "morphology hat transform". Contour-let transform has been used for decomposition of source images. The "Otsu thresholding based weighted fusion rule" has been introduced in this algorithm in order to combine "low-frequency coefficients" for improving the visual quality performance of fused image. In order to improve the edge information of final fused image, here "high frequency coefficients" are integrated by "max fusion rule". Finally, fused image reconstruction is done by using inverse contour-let transform. Proposed method has produced much better results than latest similar existing methods both in visual quality comparison and also in metric-value comparison.