An Infrared Image Enhancement Method Combining Double-Threshold Neighborhood Histogram Equalization

This study addresses the issues of insufficient image contrast, blurred target details, and over-enhancement faced by traditional infrared image enhancement algorithms. We propose an infrared image enhancement method that combines double-threshold neighborhood histogram equalization. First, the proposed method uses a bilateral filter to decompose the infrared image to be corrected into a base layer and a detail layer. Then, the base layer undergoes gray level compression and contrast enhancement using a combination of double-threshold neighborhood histogram equalization and contrast-limited histogram equalization. Subsequently, the detail layer is enhanced using weighted guided filtering. Finally, the processed results of each layer are fused through scientifically determined weight factors, resulting in an enhanced infrared image. Experimental results show that, compared to the traditional CLAHE algorithm, the proposed method achieves an improvement of up to 5.3% in the information entropy index and up to 3.6% in the SSIM index.