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

Maritime ship detection method and system based on low-rank representation and sparse constraint of adaptive weight nuclear norm approximation

An adaptive weight and sparse constraint technology, applied in the field of remote sensing image processing, can solve the problems of misjudgment of details and false alarm targets in detection results.

Pending Publication Date: 2021-07-06
NANJING UNIV OF SCI & TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The abnormal object detection algorithm based on the traditional low-rank representation and sparse constraints, because the weight of each singular value is treated equally in the solution process, often in the case of a complex background, it will cause misjudgment of some details of the background, Causes more false alarm targets in the detection results

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
  • Maritime ship detection method and system based on low-rank representation and sparse constraint of adaptive weight nuclear norm approximation
  • Maritime ship detection method and system based on low-rank representation and sparse constraint of adaptive weight nuclear norm approximation
  • Maritime ship detection method and system based on low-rank representation and sparse constraint of adaptive weight nuclear norm approximation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] combine figure 1 , the present invention is based on the low-rank representation of adaptive weight nuclear norm approximation and the sea ship detection method of sparse constraints, the specific process is:

[0021] Step 1, perform adaptive background learning on the background training samples according to the improved singular value decomposition algorithm:

[0022] For background training samples y B ∈ R m×n , m and n represent the number of matrix rows and columns, the background dictionary is calculated according to the mathematical model, the formula is as follows:

[0023]

[0024] In the formula, A B is the background dictionary, γ is the sparse vector, and K is the degree of sparsity.

[0025] Since the above problem has two variables that need to be optimized, it is generally solved by fixing one of the variables, optimizing the other, and then alternately fixing and updating the optimization. Therefore, the problem is decomposed into two parts, spar...

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 maritime ship detection method and system based on low-rank representation and sparse constraint of adaptive weight nuclear norm approximation. The method comprises the following steps: carrying out adaptive background learning on a background training sample according to an improved singular value decomposition algorithm; modeling the hyperspectral ocean image data according to low-rank representation and a sparse constraint model; when the low-rank background matrix is recovered, updating the weight according to the singular value of the last iteration, so that the purpose of adaptive weight nuclear norm approximation is achieved; and processing the hyperspectral ocean image data to obtain a sparse matrix, and detecting a ship target according to a detection operator. According to the invention, rapid and accurate ship detection can be carried out on hyperspectral ocean image data.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, in particular to a marine ship detection method and system based on low-rank representation and sparse constraints of adaptive weight kernel norm approximation. Background technique [0002] Hyperspectral remote sensing technology, also known as imaging spectroscopy, has gradually become a distinctive cutting-edge remote sensing technology since its birth in the early 1980s. It uses airborne or spaceborne sensors to collect long-distance radiation information of ground objects. The electromagnetic wave characteristics of different ground objects are different, so it can be used as a basis for distinguishing different ground objects. Hyperspectral remote sensing is based on the traditional multispectral remote sensing technology, further improving its spectral resolution. Usually multispectral data consists of 5 to 10 bands with a bandwidth of 70-400nm, while hyperspectral...

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): G06F30/20G06F111/04
CPCG06F30/20G06F2111/04
Inventor 袁飞吴泽彬徐洋韦志辉陆威
Owner NANJING UNIV OF SCI & TECH