为了改善热红外图像的增强效果，本文提出了一种基于改进的直方图裁剪方法的热红外图像增强算法。该算法核心是确定热红外原始图像与传统的直方图均衡化图像的直方图 bins中像素点的数量差，再根据范围准则，将计算出的不同 bins的差值划分为不同的区块。然后重新分配直方图，确定变换函数，得到增强后的热红外图像。该算法是一种改进的全局直方图均衡化方法，在对比度增强、直方图形状和细节信息之间可以做到较好的平衡。本文算法的峰值信噪比、结构相似度和均方误差的平均值分别为 27.5、0.923和 0.59，均优于其他算法，实验结果表明，该方法能有效增强热红外图像。

To improve the enhancement of infrared thermal images, this paper proposes an algorithm based on an improved histogram clipping method. The algorithm determines the difference between the number of pixels in the histogram bins of the original thermal infrared image and the traditional histogram equalization image. Subsequently, based on a specific range criterion, the calculated difference of various histogram bins is divided into different blocks. Next, the histogram is redistributed, and a transform function is determined to obtain the enhanced thermal infrared image. The algorithm is based on an improved global histogram equalization method, which achieves a suitable balance between contrast enhancement, histogram shape, and detail information. The experimental results demonstrate that the average values of the peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and mean-square error (MSE)—27.5, 0.923, and 590 respectively are better than those of other algorithms. Thus, this method effectively enhances the thermal infrared image.