SAR image segmentation with parallel region merging
- PDF / 6,569,991 Bytes
- 21 Pages / 439.642 x 666.49 pts Page_size
- 3 Downloads / 227 Views
SAR image segmentation with parallel region merging Zejun Zhang1 · Xiong Pan2,3 · Kai He2,3 · Li Cheng4 · Changcai Yang3 · Riqing Chen3 Received: 4 April 2020 / Revised: 10 August 2020 / Accepted: 16 September 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In this paper, a parallel region merging strategy is proposed for partitioning Synthetic Aperture Radar (SAR) images into several disjoint regions, based on the region adjacency graph (RAG) of an initial partition and the nearest neighbor graph (NNG) produced from the RAG. Developed from the multi-direction ratio edge detector, a multi-scale-multi-direction (MSMD) one is used to extract edge strength map (ESM) of an initial SAR image, feeded into watershed transform to generate an initial partition result of the initial SAR image. Considering local image properties, which makes the generated NNG center around binode circles situated in interiors of homogeneous regions, many of which are independently located in different homogeneous regions, the predication of the parallelizability for bi-node circles is proposed to make the proposed parallel region merging strategy. The proposed parallel merging strategy simultaneously merges bi-node circles far away from boundaries of regions, characterized by the length of path from a node to the bi-node circle in the NNG. The performance of the proposed parallel merging strategy is analyzed theoretically and experimentally, and our experiments show that the proposed method outweighs other compared methods. Keywords Synthetic Aperture Radar (SAR) images segmentation · Multi-scale-multi-direction (MSMD) ratio edge detector · Parallel region merging strategy
1 Introduction Synthetic Aperture Radar (SAR) images segmentation technology is a fundamental skill for the scene interpretation [16], which provides the structural information of the scene by Zejun Zhang
zjzhang [email protected] 1
College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua, 321004, China
2
Key Laboratory of Smart Agriculture and Forestry, Fujian Province University, Fujian Agriculture and Forestry University, Fuzhou, 350002, China
3
The Digital Fujian Institute of Big Data for Agriculture and Forestry, College of Computer and Information Sciences, Fujian Agriculture and Forestry University, Fuzhou, 350002, China
4
Jinshan College, Fujian Agriculture and Forestry University, Fuzhou, 350002, China
Multimedia Tools and Applications
partitioning the SAR image into several disjoint homogeneous regions. Because of the existence of the strong speckle noise in SAR images yielded by the coherent imaging principle of microwave imaging technology, many segmentation methods effective for optical images can not be used directly in the field of the SAR image segmentation, and it also difficult to simply expand these methods to satisfy SAR image segmentation. One kind of the major SAR image segmentation methods is the region merging based technique, which iteratively merges the most similar
Data Loading...
