Adaptive new top-hat transform and multi-scale sequential toggle operator based infrared image enhancement

The contrast enhancement, brightness preserving and natural look of the output image plays an important role in an infrared (IR) image enhancement process. There are several IR image enhancement algorithms that have been developed, but they may produce artificial looking images, and the images derived from these methods are not suitable for applications such as surveillance, environmental monitoring and so on. In this paper, an algorithm using new top-hat transform and multi-scale sequential toggle operator has been proposed. Firstly, a new white top-hat transform and new black top-hat transform is generated to extract the light and dark regions of the image. Then, the multi-scale sequential toggle operator is used on the extracted light and dark regions to get a sequence of regions. Ultimately, the final image is extracted from a sequence of images. To test the proposed method, the experiment is carried out on the OSU thermal bench mark database images. The enhanced image results have been compared with the popular existing methods such as Histogram Equalization (HE) and Adaptive Histogram Equalization (AHE), max-median (3 × 3) using various evaluation measures. From the results, it is observed that the proposed method works better than the existing methods.