Abstract: Considering the characteristics of the inconspicuous difference between targets and backgrounds and the low contrast in infrared images，an adaptive enhancement algorithm based on fuzzy particle swarm optimization is used in the infrared image processing.The gray transformation is better one of the infrared image enhancement methods，but the suitable threshold is powerful guarantee for obtaining the good enhancement effect.Under the maximal entropy criterion，the auto-adapted threshold is selected by particle swarm optimization.Then using fuzzy gray transformation，the infrared image gray is auto-adapted stretched and image is enhanced.Experiments show that compared with histogram equalization，the presented algorithm can reduce the background influence over the goal and improve the contrast of infrared imagery.

摘要： 针对红外图像目标与背景区分不明显、对比度低的特点，把粒子群优化算法应用到红外图像增强中，提出了红外图像自适应增强的模糊粒子群优化算法。灰度变换增强是红外图像增强的首选方法之一，而选取适当的阈值是其取得良好的增强效果的有力保证。该算法通过粒子群优化算法来寻求最大熵准则下的自适应阈值，然后用模糊灰度变换增强方法自适应地拉伸红外图像灰度，增强图像。仿真实验表明，相对于常见的直方图处理，该算法能降低红外图像中背景对目标的影响，能提高红外图像的对比度。