论文部分内容阅读
针对SAR影像进行水上桥梁目标检测问题,提出了一种纹理特征与关联特征相结合的方法。该方法首先通过去噪及增强预处理改善图像质量,并采用阈值分割方法快速提取初始目标;然后使用灰度共生矩阵对目标及其周边背景纹理进行差异对比分析,从而去除虚警目标。实验结果表明,该方法能够准确地检测出水上桥梁目标。
Aiming at the problem of target detection of waterborne bridge in SAR image, a method of combining texture features and correlation features is proposed. Firstly, the image quality is improved by denoising and enhancement pretreatment, and the initial target is quickly extracted by using threshold segmentation method. Then the gray background co-occurrence matrix is used to compare and contrast the target and the surrounding background texture to remove false targets. Experimental results show that this method can accurately detect the target of water bridge.