A skeleton detection system and detection method for noisy images

A detection system and skeleton technology, applied in the field of computer vision, can solve problems such as poor robustness, and achieve the effect of easy implementation, simple structure, and good robustness advantages

Active Publication Date: 2022-07-12
XI AN JIAOTONG UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the technical problems existing in the prior art, the present invention provides a skeleton detection system and detection method oriented to noise images to solve the problem of poor robustness when processing a large number of noise images in the prior art, and the extracted skeleton images contain many abnormal Related information on technical issues

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
  • A skeleton detection system and detection method for noisy images
  • A skeleton detection system and detection method for noisy images
  • A skeleton detection system and detection method for noisy images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] A noise-oriented skeleton detection system according to the present invention includes eight skeleton filters, each skeleton filter includes a pair of two-dimensional Gabor-like filters with mutually opposite directions; each Gabor-like filter It consists of a pair of positive and negative Gaussian filters, where the positive Gaussian filter is located in the center position, and the negative Gaussian filter is located in the surrounding position, which ensures that each skeleton filter has reflection symmetry; The absolute value is uniformly normalized, so that the sum of positive and negative in each skeleton filter is zero, which ensures that each skeleton filter has a zero-sum structure.

[0055] Eight skeleton filters are respectively set in eight directions of the x-y plane of the Cartesian coordinate system, wherein the skeleton filters in seven directions are obtained by rotating the skeleton filters in one direction in the x-y plane of the Cartesian coordinate s...

Embodiment 2

[0068] The principle of Embodiment 2 is basically the same as that of Embodiment 1, the difference is that in Embodiment 2, the positive Gaussian filter is located at the center position, and the negative Gaussian filter is located at the surrounding position.

[0069] When using the skeleton detection method described in Example 1 for noise image detection, the obtained skeleton feature image is as follows Image 6 As shown, the noise interference in the noise image to be detected can be handled very robustly, and it is suitable for the target object brightness in the noise image to be detected is lower than the background image.

Embodiment 3

[0071] When there are parts of the same noise image to be detected that the brightness of the target object is higher than that of the background image, and there are also parts of the target object that are lower than the background image, the noise image to be detected is processed by using Embodiment 1 and Embodiment 2 respectively, The results obtained in Embodiment 1 and Embodiment 2 are combined by an addition operation, so as to realize the processing of noise images with different brightness of target objects in the noise images to be detected.

[0072] When using the skeleton detection method described in Example 3 for noise image detection, the obtained skeleton feature image is as follows Figure 7 As shown, the noise interference in the noise image to be detected can be handled very robustly, and is suitable for the noise image to be detected in various situations.

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 noise image-oriented skeleton detection system and detection method, comprising a plurality of skeleton filters, each skeleton filter has a zero-sum structure and a reflection symmetry structure; the plurality of skeleton filters are respectively arranged in the In the n directions of the x-y plane of the Karl coordinate system, a plurality of skeleton filters are distributed with rotational symmetry; the invention provides that the filtering of uniform noise distribution is realized by adopting the skeleton filter with zero-sum; reflection symmetry is adopted; The skeleton filter of the skeleton filter realizes the filtering of the noise distribution without a symmetric structure; multiple skeleton filters are set to be rotationally symmetrical, and the center of the connecting line of the two opposite directions inside each skeleton filter is the skeleton point, and then realize It has a simple structure, is easy to implement, and has a very good robustness advantage for noise image skeleton extraction.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a noise image-oriented skeleton detection system and detection method. Background technique [0002] The skeleton of an object is an abstract representation method based on the structure of an object, which stores each point at the central axis where the boundaries of the concrete object are equal, and the distance is the radius of the relevant inner sphere; the skeleton feature can explain the structural information of the object and the various components at the same time. It has extensive research in the field of computer vision. For deformable objects, skeleton features provide a concise and effective low-dimensional representation method, and can be applied in a variety of scenarios, such as: object recognition and retrieval, natural scene text detection, pose estimation, shape matching, character localization, Medical diagnosis and lane line detection and other fie...

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 & AuthorityPatents(China)
IPC IPC(8): G06V20/10G06V10/30
CPCG06V20/00G06V10/30
Inventor白秀秀叶乐乐杨明刘哲
OwnerXI AN JIAOTONG UNIV