Circular Trace Transform and Its Applications to Image Texture Analysis

WANG Yu-ling; LI Ming

doi:10.3969/j.issn.0372-2112.2018.10.007

您当前的位置：

首页 >

文章列表页 >

Circular Trace Transform and Its Applications to Image Texture Analysis

更新时间：2025-07-16

- Circular Trace Transform and Its Applications to Image Texture Analysis
- Acta Electronica Sinica Vol. 46, Issue 10, Pages: 2351-2358(2018)
- 作者机构：
  
  1. 南京航空航天大学自动化学院,江苏,南京,211106
  2. 东华理工大学江西省放射性地学大数据技术工程实验室,江西,南昌,330013
  3. 南昌航空大学江西省图像处理与模式识别重点实验室,江西,南昌,330063
  4. 南京航空航天大学自动化学院,江苏,南京,211106
  5. 东华理工大学江西省放射性地学大数据技术工程实验室,江西,南昌,330013
  6. 南昌航空大学江西省图像处理与模式识别重点实验室,江西,南昌,330063
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.61262019);Program of Advantage Sci-Tech Innovative Team of Jiangxi Province (No.20113BCB24009);Science and Technology Program of Education Department of Jinagsu Province (No.GJJ160554, No.GJJ170432);Open Fund for Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition (No.ET201880042);Open Foundation for Engineering Laboratory of Radio-geonomy Big Data Technology of Jiangxi Province (No.JELRGBDT201707)
- DOI：10.3969/j.issn.0372-2112.2018.10.007
  CLC： TP391.4TN911.73
- Published Online：25 October 2018，
  
  Published：2018
- 稿件说明：
移动端阅览
WANG Yu-ling, LI Ming. Circular Trace Transform and Its Applications to Image Texture Analysis[J]. Acta Electronica Sinica, 2018, 46(10): 2351-2358.
DOI：

WANG Yu-ling, LI Ming. Circular Trace Transform and Its Applications to Image Texture Analysis[J]. Acta Electronica Sinica, 2018, 46(10): 2351-2358. DOI： 10.3969/j.issn.0372-2112.2018.10.007.

摘要

针对迹变换提取的特征对于含圆弧形纹理图像描述欠佳的问题，提出了圆迹变换及其纹理特征提取方法.圆迹变换使用圆迹线来跟踪图像，并将圆迹线上的泛函结果映射到由半径、长度和角度三个参数生成的空间上，对该结果继续做泛函积分，能得到图像四重圆迹空间特征.在圆迹线上使用不同的泛函能得到不同的圆迹变换结果，不同的泛函组合能获得不同的圆迹变换特征，能表示图像不同的纹理性质和更深层次的图像信息，产生多维的圆迹变换纹理特征信息.在Coil-20对象图库及Brodatz纹理库上对含圆弧形图像的分类实验表明，在训练样本较少的情况下，利用圆迹变换提取的纹理特征，其识别能力比迹变换都有明显提升，对于包含圆形、弧线形纹理的图像具备了更好的分辨能力.

Abstract

Circular trace transform (CTT) is proposed to extract the texture features that are more suitable for describing images containing circular or arc-shaped texture than the ones extracted by Trace Transform. CTT consists of tracing an image with circles around which certain functionals of the image function are calculated. The functional results on the circular trace are mapped to the space generated by three parameters such as radius

length and angle. With functional integral in the results

the image quadruple texture features can be generated. Different circular trace transformations can be obtained by using different compositions of functionals and the quadruple features by CTT can represent different texture properties and deeper intrinsic information of images. It shows better performance on recognizing images with circular or arc-shaped texture in Coil-20 and Brodatz database. In the case of fewer training samples

the recognition capability of the features is obviously improved.

关键词

Keywords

references

Views

214

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on the Generation of Adversarial Samples Based on the Attribution of Principal Features

Research on Dynamic Continuous Emotional Recognition of EEG Based on Improved TCNN Algorithm

Discriminative Category Prompt Learning Based on Image Content Understanding

Research on Lightweight Sensing Technology Based on Single Antenna and Multiple Subcarriers

Related Author

WANG Shuo

XU Ru-zhi

GUAN Zhi-tao

JIE Li-lin

LIU Yong

WANG Ming-xun

ZOU Yang-meng

XU Yi-lu

Related Institution

School of Control and Computer Engineering， North China Electric Power University

School of Instrument Science and Optoelectronic Engineering, Nanchang Hangkong University

School of Aviation Services and Music, Nanchang Hangkong University

School of Software, Jiangxi Agricultural University

School of Aeronautical Manufacturing and Mechanical Engineering, Nanchang Hangkong University

⁰