论文部分内容阅读
高分遥感影像分割质量对面向对象分类精度有着重要的影响,良好的影像分割质量有利于得到较高的分类精度。对高分遥感影像分割质量进行评价,从而找到最优的分割结果就显得十分重要。通过对比参考对象和分割对象之间的面积和位置差异,提出了一种新的高分遥感影像分割质量评价方法。将该评价方法应用于GeoEye-1高分遥感影像分割质量评价,试验结果表明:该评价方法能客观地评价影像分割质量,所得到的最优分割结果与参考对象边界匹配程度高,有利于影像后续的分类。
High-resolution remote sensing image segmentation quality has an important impact on object-oriented classification accuracy, good image segmentation quality is conducive to higher classification accuracy. It is very important to evaluate the quality of high-resolution remote sensing image segmentation to find the optimal segmentation result. By comparing the area and position difference between the reference object and the segmentation object, a new method for quality assessment of high resolution remote sensing image segmentation is proposed. The evaluation method is applied to the quality evaluation of GeoEye-1 high-resolution remote sensing image segmentation. The experimental results show that this method can objectively evaluate the quality of image segmentation, and the optimal segmentation result matches the reference object boundary, which is in favor of image Subsequent classification.