Target recognition method and device

A target recognition and target technology, applied in the field of image processing, can solve problems such as insufficient ability, strong environmental noise and system noise, complex background, etc., to prevent misjudgment and suppress system noise and environmental noise.

Inactive Publication Date: 2015-10-07
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF5 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, the power line is a very weak linear object in remote sensing images, which has the characteristics of small dimension (at the sub-pixel level), complex background, etc.
When using the above-mentioned traditional technology to identify this kind of weak linear target, due to the presence of environmental noise and system noise that are stronger than the target to be identified, various problems such as misjudgment often occur
Furthermore, existing linear object recognition methods are often insufficient when dealing with short line segments

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
  • Target recognition method and device
  • Target recognition method and device
  • Target recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0066] Figure 4 (a) shows a synthetic image with a real satellite image as the background, which includes forests, buildings, roads, etc. There are two power lines from the upper left to the lower right in the image, and the two roofs in the left middle and upper right corner of the background of the image are high-brightness planar targets. Figure 4 (b) shows the results of linear feature detection using traditional Radon transform, where almost all false straight lines are caused by bright planar objects. Figure 4 (c) The result of linear feature detection using traditional Radon transform after edge detection. From Figure 4 (c) It can be seen that although the false straight lines caused by bright planar targets are successfully eliminated, there are a large number of false straight lines caused by forest land in the detection results. Figure 4 (d) shows the result of linear feature detection using the method of the present invention after edge detection. From Fi...

example 2

[0068] Figure 5 The composite image of (a) includes short line segments to simulate the wake of a ship, and this V-shaped wake is found to be useful for tracking the ship. Gaussian noise pair with variance equal to 5 Figure 5 (a) Perform noise processing to get Figure 5 (b) The image to be processed. Using the target recognition method proposed by the present invention to Figure 5 (b) After target recognition, get the following Figure 5 (c) The results shown. From Figure 5 (c) It can be seen that the method of the present invention can also successfully identify short line segments. Figure 5 (d) and Figure 5 (e) respectively show the two-dimensional distribution in Radon space when using traditional Radon numbering to perform linear feature detection and using the method of the present invention to perform linear feature detection. By comparing the two figures, it can be seen that the method of the present invention effectively suppresses the noise and enhances...

example 3

[0070] Figure 6 (a) is the original image to be processed, which is a GeoEye-1 remote sensing image with a resolution of 512×512. From Figure 6 In (a), it can be seen that there is a power line diagonally across the image from the north to the southeast on the remote sensing image. The background is very complex, including bare land, roads, substations, and large forests in the south. There is a lot of noise, which increases the identification and recognition of power lines. extraction difficulty. The result of edge detection is as follows Figure 6 As shown in (b), it can be seen that a large amount of system noise exists. Figure 6 (c) shows the result of target recognition using the method of the present invention. From Figure 6 It can be seen in (c) that in addition to the power lines of interest, the road information of the image to be processed is also extracted. At this time, in order to further extract the power lines to be extracted, the following characteris...

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 target recognition method and a device. The target recognition method comprises the following steps: a linear feature detection method is used for carrying out linear feature detection on a to-be-recognized image, detected linear features are obtained, and according to the linear feature detection method, detection on image space linear features is converted into detection on an another space extreme point, and a continuous clustering operator is used for enabling a continuous image point to have a larger weight than a discrete image point in the conversion process; and according to features of a to-be-recognized target in the to-be-recognized image, the to-be-recognized target is recognized in the detected linear features. By using the target recognition method and the device of the invention, a weak linear target can be detected and recognized in a high-resolution remote sensing image, system noise and environment noise can be effectively restrained, the target is recognized and extracted successfully, erroneous judgment is prevented, and a short line segment can be successfully detected.

Description

technical field [0001] The present invention relates to image processing technology, in particular, to a linear target recognition method and device. Background technique [0002] Remote sensing images include rich surface linear features, such as roads, railways, rivers, coastlines, etc. Recognizing linear features from remote sensing images is a problem that has been widely concerned by researchers. Considering the importance of road information and the need to update the earth information system, past researches mostly focused on road information extraction. In the past few decades, many researchers have proposed various solutions to this problem to identify and extract road information from high-resolution aerial images, Synthetic Aperture Radar (SAR) images, and high-resolution satellite images. [0003] Another important target in remote sensing images is the power line. Traditional research focuses on extracting power line information from LiDAR images, helicopter ...

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): G06K9/00G06K9/46G06V10/44G06V10/48
CPCG06V20/13G06V10/48G06T7/13G06V20/182G06V10/255G06V10/44G06T2207/10032G06T2207/30184G06V20/176
Inventor 陈云坪童玲韩威宏
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap