基于遥感影像的园林水景区域目标提取仿真

Target Extraction and Simulation of Garden Waterscape Area Based on Remote Sensing Image

提出一种新的园林水景区域目标提取方法。利用暗通道优先去雾方法对园林遥感影像展开增强处理;根据遥感影像的相似性权值将其划分为平均图像和对比图像,引入滤波器分别对上述图像展开滤波处理,获得去噪后的园林遥感影像;结合K最近邻和局部二值模式对处理后的遥感影像展开水陆粗分离处理。通过支持向量机分类器细分离遥感影像的水陆区域。基于此,利用形态学腐蚀运算对水景区域边缘展开平滑处理,提取园林的水景区域。实验结果表明,研究方法应用下图像可完成显著增强,去噪效果明显,对景观的水体目标提取精度较高。

In this paper,a method of target extraction for landscape waterscape region was put forward.Firstly,the dark channel priority defogging method was a dopted to enhance the remote sensing image of the landscape.According to the similarity weight,the remote sensing image was divided into the average image and the contrast image.Then,a filter was introduced to filter the above ima ges respectively and thus to obtain the denoi sed images.The K-nearest neighbor was combined with the local binary to perform the rough separation of water and land on the remote sensing image.Moreover,the Support vector machine classifier was used to separate the land and water regions in remote sensing images.On this basis,the edge of the waterscape area was smoothed by morphological erosion operation.Finally,the waterscape region of the garden was extracted.Experimental results show that the image is significantly enhanced and the denoising effect is obvious under the application of the method,and the extraction accuracy for the water target is higher than before.