Detection method of main slip line of river channel based on sne manifold learning
A technology of manifold learning and detection methods, applied in image analysis, image enhancement, instruments, etc., can solve problems such as poor detection accuracy, improve detection accuracy and robustness, and overcome feature uncertainty.
Inactive Publication Date: 2018-11-13
黄河水利委员会信息中心 +1
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In order to overcome the shortcomings of poor detection accuracy of the existing main channel line detection method, the present invention provides a method for detecting the main channel line based on SNE manifold learning
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
[0038] refer to Figure 1-2 . The present invention is based on the SNE manifold learning method for detecting the main slip line of the river. The specific steps are as follows:
[0039] 1. Segment the river and generate a river segmentation image.
[0040] On a TM remote sensing image, select band5, which is most sensitive to water bodies, for rough segmentation of rivers. Using the region growing method, the river is grown by selecting the seed point pixels and defining the similarity measure to generate a binary image of the same size as the original image, in which the value of the river region is 1, and the value of the non-river region is 0 . Then use the generated binary image to mask the original image to obtain the segmented remote sensing image. The pixel value of the river part in the image is the spectral value of the original image, and the non-river part is all 0.
[0041] 2. Extract the channel centerline.
[0042] On the binary image generated above, the ...
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 invention discloses a method for detecting a main channel line of a river channel based on SNE manifold learning, which is used to solve the technical problem of poor detection accuracy of the existing method for detecting a main channel channel of a river channel. The technical solution is to construct a 3×3 window with a single pixel as the center, and use the spectral value of the pixel at the center of the window, the statistical value of the window gray level, and the window gradient to construct a 24-dimensional feature vector to describe the main feature. Use manifold learning to reduce the 24-dimensional feature space to 2-dimensional space, and then seek the point closest to the known main chute pixel in 2-dimensional space, mark this point as the downstream main chute position, and so on until the entire River section detection is over. This method uses windows to construct high-dimensional feature vectors of the main stream, uses manifold learning to reduce the dimension, and can seek the low-dimensional realization of the point closest to the known main stream point in the constructed high-dimensional feature space. It overcomes the technical problem of feature uncertainty in the detection of the main chute, and improves the detection accuracy of the main chute.
Description
technical field [0001] The invention relates to a method for detecting a main channel line of a river course, in particular to a method for detecting a main channel line of a river channel based on SNE manifold learning. Background technique [0002] The main chute is the line connecting the points with the highest vertical flow velocity on the cross-section of the river channel. It is a generalized representation of the main chute of the river channel and is used to describe the change of the river regime. The traditional acquisition method is to determine the position of the main chute by manually checking the river by boat, and then draw and transfer it manually on the topographic map. This method often takes a long time, has poor safety, and it is difficult to obtain the change of the main chute during the flood period . At present, there are no foreign reports on the application of remote sensing images to interpret the main channel of the river, and domestic research ...
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): G06T7/00
CPCG06T2207/10032G06T2207/20081G06T2207/30181
Inventor 韩琳刘学工张艳宁佘红伟邓刚
Owner 黄河水利委员会信息中心