低空目标的图像灰度相似度投影检测方法

Image gray similarity projection detection method for low-altitude target

为了解决传统方法检测低空目标时背景抑制效果较差的问题,提出一种低空目标的图像灰度相似度投影检测方法。通过图像的行灰度相似度投影曲线数值跳变位置分割出天空区域,移除地面复杂背景对天空区域目标检测的影响。依据目标与云朵和天空背景的灰度值差别,对像素点灰度相似度行和列投影曲线进行综合分析来确定目标的个数。由极值点结合图像分割确定目标的行和列像素坐标从而对目标进行定位。实验结果表明,该方法能有效检测拥有复杂地面背景和环境光照变化下的低空目标,与传统方法相比,检测速度提高了14%,误检率明显改善,表现出较好的背景抑制效果。系统实时性强,适用于低空小目标检测应用,为后续研究提供了有效的技术手段。

In order to solve the problem that the traditional method has poor background suppression effect when detecting low-altitude targets,an image gray similarity projection detection method for low-altitude target is proposed.The sky area is segmented by the numerical jump position of the row gray similarity projection curve of the image,and the influence of the complex ground background on the target detection in sky area is removed.According to the difference in gray value between the target and cloud,sky background,the number of targets is determined by the comprehensive analysis of the row and column projection curves of pixel gray similarity.The row and column pixel coordinates of the target are determined by extreme points combined with image segmentation to locate the target.The experimental results show that this method can effectively detect low-altitude targets with complex ground background and illumination change of environment.Compared with the traditional method,the detection speed is increased by 14%,while the false detection rate is significantly improved,showing better background suppression effect.The system has strong real-time performance,which is suitable for low-altitude small target detections,and provides an effective technical means for follow-up research.