Infrared image enhancement using improved contrast-limited adaptive histogram equalization

Infrared images often face issues like noise, low contrast, and blurred details, which limit their effectiveness in military and civilian applications. While contrast-limited adaptive histogram equalization (CLAHE) is a useful method for improving image contrast, it and related improved methods have drawbacks such as block artifacts, noise amplification, and loss of details. This paper introduces an improved CLAHE method to solve these problems. First, guided image filtering (GIF) is used to separate the image into base and detail layers, utilizing its edge-preserving ability to reduce noise and maintain edges. Then, the base layer is enhanced with an improved CLAHE method that adapts clipping limits based on the image’s dynamic range and entropy. The method also combines global and local mapping to eliminate block effects. Finally, the enhanced base and detail layers are merged to produce the improved image. Experimental results show that this approach effectively increases contrast while reducing noise and block artifacts, outperforming AHE, CLAHE, and FC-CLAHE in enhancement quality.