An improved seeded region growing algorithm
Pattern Recognition Letters
IEEE Transactions on Pattern Analysis and Machine Intelligence
Mustererkennung 1995, 17. DAGM-Symposium
Remote Sensing, Third Edition: Models and Methods for Image Processing
Remote Sensing, Third Edition: Models and Methods for Image Processing
Automatic image segmentation by integrating color-edge extraction and seeded region growing
IEEE Transactions on Image Processing
| Hi-index | 0.00 |
The Loess Plateau located in the upper and middle reaches of the Yellow River is the most serious soil erosion region in China. Check-dams (also known as silt trappers) have been demonstrated to be an effective soil and water conservation measure in the region and a traditional workflow for check-dam planning and engineering is time-consuming and cannot meet the requirement of efficiently locating optimal check-dam sites. Fine resolution satellite imagery and analysis can play a key role in screening and determining special river bed segments that can be candidates for check-dam sites. In this research, HongShiMao watershed of the Loess Plateau was selected as our case study area. Based on a detailed analysis of spectral characteristics of a fused SPOT-5 imagery for dominant land covers and geomorphological features of a constructed digital elevation model, the river bed of key channels within the watershed was automatically identified. Then we selected four check-dam sites on the river bed and four orientations of a check-dam site to explore locational and directional profiles. Such profile information is most useful for locating optimal check-dam sites in a cost-effective manner and reducing associated expenditures surrounding check-dam constructions. This remote sensing application demonstrates the latest spatial information technologies such as fine resolution satellite imagery and 3D geospatial visualisation hold promises for changing traditional workflows and advancing scientific decision making of environmental conservation projects.