A Sea-Land Segmentation Method for Visible Light Remote Sensing Images Based on Graph Segmentation and Supervised Learning
A remote sensing image, sea and land segmentation technology, applied in the field of visible light remote sensing images, can solve the problem of difficult separation of ocean and land, and judgment as ocean, etc., to achieve good overall effect, ensure correctness, and ensure accuracy.
Active Publication Date: 2018-11-09
BEIHANG UNIV +1
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The current traditional sea and land segmentation methods such as Otsu threshold segmentation, Bayesian segmentation of maximum likelihood, and statistical model segmentation based on the sea, etc., are difficult to accurately separate the ocean and land in visible light remote sensing images, especially when some When the grayscale information of the area is similar to that of the ocean, these methods can easily judge the similar area on the land as the ocean
Method used
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
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0048] In order to better understand the technical solution of the present invention, the embodiments of the present invention will be further described below in conjunction with the accompanying drawings:
[0049] The present invention is realized under the programming environment of Visual Studio 2010 and MATLAB 2014b. The main sea-land segmentation process is completed in the VisualStudio 2010 programming environment, and the linear SVM classifier training is completed in the MATLAB 2014b programming environment. After the computer reads the visible light remote sensing image, it first performs sea and land segmentation, divides the image into several regions, and then extracts the statistical features of each region (full variation, gray-scale direction histogram and gradient direction histogram) to train linear SVM Classifier, and use the trained classifier to judge whether each area is ocean or land based on the graph segmentation, and finally complete the sea-land segme...
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 PUM
Login to View More Abstract
The present invention is a sea and land segmentation method for visible light remote sensing images based on graph segmentation and supervised learning, which has five major steps: Step 1: Computer reads data; Step 2: Image segmentation; Step 3: Extracting image regions after segmentation Statistical features; step 4: train a linear SVM classifier; step 5: use the trained linear SVM to judge sea and land, and obtain the final sea and land segmentation result map. The present invention overcomes the deficiencies of the prior art, well solves the problem of sea and land segmentation in visible light remote sensing images, and achieves better segmentation results, so this method can be applied to sea and land segmentation, one of the ship detection processes of visible light remote sensing images , has broad application prospects and value.
Description
Technical field: [0001] The invention relates to a sea-land segmentation method for visible light remote sensing images based on graph segmentation and supervised learning, and belongs to the technical field of visible light remote sensing images. Background technique: [0002] Remote sensing technology refers to the application of optical cameras, radar and other detection instruments to record the electromagnetic wave characteristics of ground or space targets from a distance without physical contact, and to analyze and determine the characteristics of the target. Films or photos that record the characteristics of electromagnetic waves collected by various detection instruments (ie sensors) are collectively referred to as remote sensing images. Different types of sensors can record and obtain electromagnetic wave signals in different bands reflected by objects. According to the band distribution range of the electromagnetic wave signals recorded and obtained, remote sensin...
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 Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/34G06K9/62
CPCG06V20/13G06V10/267G06F18/2411
Inventor 史振威雷森张璐吴俊
Owner BEIHANG UNIV