遥感英语储备2

最新推荐文章于 2024-08-11 19:00:59 发布

原创最新推荐文章于 2024-08-11 19:00:59 发布 · 761 阅读

4 ·

CC 4.0 BY-SA版权

遥感论文专栏收录该内容

19 篇文章

订阅专栏

本文探讨了高分辨率遥感图像的分割过程及其面临的挑战，如迭代成本高、大数据处理能力不足等问题，并介绍了Mean Shift算法作为一种非参数聚类技术的优势及优化方案。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

1、high resolution remote sensing image segmentation process 高分辨率的遥感图像分割过程

2、one bottleneck of the mean shift procedure is the cost per iteration, especially in the huge data 瓶颈是迭代的开销，特别是大数据的开销

3、 iteration times迭代次数 adaptive adjustment 自适应调整 high resolution mass data 高分辨率大数据

use pixel as the basic unit of recognition data 应用像素作为识别数据的基本单元

4、lack of processing capacity for high resolution mass data 对于高分辨率大数据缺少处理空间

5、the spatial feature of texture of ground objects and topological relations are essentially the region level（区域层特征） characteristics ,feature extraction（特征提取） in the homogeneity region (同质地区)of the obtained image is obviously more reasonable than feature extraction in any given square window in the region.

6、mean shift algorithm is a nonparametric clustering technique which dose not require prior knowledge of the number of clusters ,and dose not constrain the shape of the clusters .

7、receive much concern high computational complexity 计算复杂度 an optimization of MS 对MS的优化

sampling points 采样数据