High dynamic range infrared image detail enhancement based on histogram statistical stretching and gradient filtering

In order to improve the image contrast and strengthen the details of high dynamic range (HDR) infrared (IR) image, a detail enhancement method based on histogram statistical stretching (HSS) and gradient filtering (GF) is proposed. First, the outliers in the HDR image are clipped by the proposed histogram statistical strategy and the clipped histogram is then extended to a new grayscale range to acquire a better contrast of view. The details in the HDR image are extracted by using the GF and its result is then adjusted by using the HSS to enhance the low-contrast detail perception. Finally, the GF result is superposed with the HSS result in a proper way to generate the final detail-enhanced image. The contribution and innovation made is threefold. A new technique for visualization of HDR image especially tailored to IR image is proposed. The effectiveness and convenience are shown by analyzing the experimental images that represent the typical and common IR scenes. Last, the performance is quantitatively assessed compared with other well-established methods. The simulation and experimental results approve its low cost, low complexity and promising outlook for real-time processing.