Watershed segmentation method based on improved fast merging algorithm

A watershed segmentation and merging algorithm technology, which is applied in computing, image analysis, image data processing, etc., can solve problems such as loss of contour information, over-segmentation, etc.
CN107945183AActive Publication Date: 2018-04-20三亚中科遥感研究所
4 Cites 7 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
三亚中科遥感研究所
Publication Date
2018-04-20

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a watershed segmentation method based on an improved fast merging algorithm. The method includes the following steps: (1) developing a complete set of preprocessing-based watershed segmentation algorithm technological process based on GF-1 data; (2) studying the influence of the global markup threshold alpha and the regional adaptive threshold adjustment coefficient g in the preprocessing-based watershed segmentation algorithm on the segmentation result, and analyzing the value ranges of alpha and g; (3) based on the initial segmentation result of the preprocessing-based watershed segmentation algorithm, using a Fast Lambda merging algorithm based on controllable common boundary length penalty to carry out region merging on the initial segmentation result through RAG and NNG to achieve a satisfactory segmentation result; and (4) studying the influence of the region merging control threshold T and the common boundary penalty coefficient lambda in the Fast Lambdamerging algorithm based on controllable common boundary length penalty on the merging result, and analyzing the value ranges of T and lambda.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to a watershed segmentation method combined with an improved fast merging algorithm, in particular, two image segmentation algorithms (watershed segmentation algorithm based on pre-processing and FastLambda algorithm based on controllable common boundary length penalty) are used to perform high-resolution remote sensing images. Image segmentation for good image segmentation results. Background technique

[0002] In remote sensing application analysis, remote sensing image segmentation is a bridge between low-level image processing and middle-level image analysis and understanding, and is a key step in completing the precise classification of high-resolution remote sensing images. In general, image segmentation algorithms include: threshold segmentation algorithm, edge segmentation algorithm and region segmentation algorithm.

[0003] For the image threshold segmentation algorithm, this method has a small amount of calculation and...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More