针对周扫红外搜索系统对空目标探测面临的图像数据量大、弱目标检测概率低、虚警率高等难点问题,提出了一种基于兴趣区(ROI)提取的目标实时检测算法。算法分析了周扫红外搜索系统获取的图像中目标与背景的特性,根据目标运动特性与灰度特性,在周扫红外搜索系统获取的整幅全景图像中快速提取目标可能存在的兴趣区;针对兴趣区内的局部目标图像切片,进一步精细检测识别,剔除虚假目标干扰。外场试验获取的实测数据目标检测结果表明,算法针对复杂低空背景下弱目标能够实现低虚警率稳健检测,已应用到了周扫红外搜索跟踪系统的工程样机研制中。 Aimed at the difficulty of detecting the empty target, such as large amount of image data, low target detection probability and high false alarm rate, an infrared target detection system based on region of interest (ROI) extraction is proposed. The algorithm analyzes the characteristics of the target and the background in the image obtained by the Zhou-bing infrared search system. According to the target motion characteristics and the grayscale characteristics, the target region of interest can be quickly extracted from the panoramic image acquired by the Zhou-bing infrared search system. Area of ​​interest within the local target image slice, further fine detection and recognition, excluding false target interference. Field test results obtained from the field data show that the algorithm can detect low false alarm rate robustly in the case of weak targets with complex low-altitude background and has been applied to the development of engineering prototypes for the infrared scanning system.