Unlock instant, AI-driven research and patent intelligence for your innovation.

Superpixel-based regional combination SAR image coastline detection algorithm

A detection algorithm and superpixel technology, applied in image enhancement, image analysis, image data processing, etc., can solve the problems of coastline detection with great difficulty and inability to merge small areas, and achieve accurate features and high edge fit. Effect

Inactive Publication Date: 2017-10-27
DALIAN MARITIME UNIVERSITY
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to factors such as coherence spots, sea breeze and the complexity of the land environment, it is difficult to detect coastlines.
When the sea surface or land is uneven, the existing coastline detection algorithm for area merging is prone to small areas that cannot be merged, and the threshold for merging needs to be artificially set

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Superpixel-based regional combination SAR image coastline detection algorithm
  • Superpixel-based regional combination SAR image coastline detection algorithm
  • Superpixel-based regional combination SAR image coastline detection algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0096] Algorithm performance comparison mainly uses root mean square error (RMSE) and QA (overall accuracy) as the accuracy analysis indicators. Firstly, RMSE is compared, and the calculation formula is as follows:

[0097]

[0098] Among them, RMSE represents the average error between the hand-painted coastline and the coastline extracted by various algorithms, x 1k Indicates the pixel value of the kth position pixel in the binary image obtained by hand-drawn coastline extraction results. x 2k Indicates the pixel value of the kth position pixel in the binary image of the coastline extraction result obtained by the above theoretical model, and N indicates the number of image pixels. The smaller the RMSE value, the closer to the real coastline and the higher the accuracy.

[0099] For Envisat images, the RMSE comparison of the algorithm is shown in Table 1.

[0100] Table 1 RMSE comparison of three algorithms for Envisat images

[0101]

[0102] According to the RMSE ...

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

No PUM Login to View More

Abstract

The invention discloses a superpixel-based regional combination SAR image coastline detection algorithm. The algorithm comprises the steps of reading an SAR image I and inputting seed points of K superpixels; calculating the positions of the seed points; calculating out the similarity Si,j between a neighborhood point mean value and a central point mean value of the seed points; then calculating a mean value and a variance of the seed points to serve as features of the seed points; repeating the steps until the types of all the points are not changed; calculating a pixel mean value of the superpixels, the number of pixels in the superpixels and a standard deviation of the pixels in the superpixels; and until the superpixels before and after iteration are no longer combined, outputting a coastline detection result. By constructing a new local window, the problem that the calculated features are fuzzy due to edges contained in a conventional rectangular window can be effectively solved; and a similarity descriptor is constructed through the local window, so that the extracted features are more accurate, and the edge fitness of the superpixels is higher.

Description

technical field [0001] The invention relates to a superpixel-based area merging SAR image coastline detection algorithm, which belongs to the field of coastline detection. Background technique [0002] Synthetic aperture radar is an active microwave detector, which uses the principle of synthetic aperture, signal processing method and pulse compression technology to synthesize a larger equivalent antenna aperture through a smaller-sized real antenna aperture for imaging. SAR images have been widely used in the identification and detection of strategic targets, disaster control, land resources monitoring, sea area use management, map surveying and mapping, ship target recognition, mineral exploration, crop growth monitoring and other fields and play an important role. In the management of sea area use, coastline detection is an important link, and changes in coastal zones can be monitored by detecting coastlines. Due to land reclamation, river sediment accumulation and other...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/13G06T7/136
CPCG06T7/13G06T7/136G06T2207/10044G06T2207/20081G06T2207/30184
Inventor 史晓非王智罡马海洋丁星冯建德刘玲
Owner DALIAN MARITIME UNIVERSITY