显著性运动目标自顶向下快速检测方法研究

Research on the top-down fast detection method of significant moving target

显著性运动目标受到动态特征扰动导致目标检测性能差,为了提高对显著性运动目标的检测能力,提出基于梯度模型边缘轮廓特征匹配的显著性运动目标自顶向下快速检测方法.先采用图像视觉特征重构方法进行显著性运动目标的图像样本采集,提取显著性运动目标的边缘轮廓特征量,然后采用梯度模型方法构建目标图像的方向梯度直方图,结合信息增强算法实现对显著性运动目标的细节特征增强处理.最后对图像进行颜色空间转换,采用梯度模型边缘轮廓特征匹配方法实现对显著性运动目标自顶向下特征重构,根据图像重构效果实现对显著性运动目标的快速检测.仿真结果表明,采用该方法进行显著性运动目标检测的准确性较高,计算开销较小,检测速度较快,提高了显著性运动目标的检测识别能力.

Significant moving targets are disturbed by dynamic features,resulting in poor target detection performance.In order to improve the detection ability of the significance moving target,a top-down fast detection method for salient moving targets based on edge contour feature matching of gradient model is proposed.The method comprises the following steps of:the image visual feature reconstruction method is used to collect the image samples of the significant moving target,the edge contour feature quantity of the salient moving target is extracted,the gradient model method is used to construct the directional gradient histogram of the target image,and the information enhancement algorithm is used to realize the detailed feature enhancement of processing of significant moving target.The color space of the image is converted,the edge contour feature matching method of the gradient model is used to realize the top-down feature reconstruction of the significant moving target,and the fast detection of the salient moving target is realized according to the image reconstruction effect.The simulation results show that the method is used to detect significant moving target with higher accuracy,less computation cost,faster detection speed,and improved detection and recognition capabilities for significant moving targets.