针对复杂背景下单帧红外图像中的弱小目标检测问题,提出了一种级联的改进Asymmetric AdaBoost算法。该算法通过提取红外小目标梯度对比度、邻域灰度分布等多个有效特征,充分利用多个红外特征的互补作用,对Asym-metric AdaBoost在仿真中出现的问题进行了相应的改进。通过对实测数据进行仿真,结果表明该方法达到了高检测概率和低虚警率,能有效地检测红外弱小目标。 Aiming at the weak target detection in single-frame infrared images with complex background, a cascaded improved Asymmetric AdaBoost algorithm is proposed. This algorithm improves the Asym-metric AdaBoost’s problems in simulation by taking advantage of multiple effective features, such as gradient contrast of infrared small target and neighborhood gray distribution, and makes full use of the complementary effect of multiple infrared features. The simulation results show that the proposed method achieves high detection probability and low false alarm rate, and can effectively detect infrared weak targets.