文献类型： 外文期刊

作者： Zhang, Xun ¹ ; Liu, Jingxian ¹ ; Zheng, Yalu ³ ; Zheng, Yan ¹ ; Hussain, Masroor ⁴ ;

作者机构： 1.Harbin Engn Univ, Coll Intelligent Syst Sci & Engn, Harbin, Peoples R China

2.Qingdao Innovat & Dev Ctr Harbin Engn Univ, Coll Intelligent Syst Sci & Engn, Qingdao, Peoples R China

3.Heilongjiang Acad Agr Sci, Inst Crop resources, Harbin, Peoples R China

4.Ghulam Ishaq Khan Inst Engn Sci & Technol, Fac Comp Sci & Engn, Topi, Khyber Pakhtunk, Pakistan

关键词： convolutional neural nets; image processing; shape recognition

期刊名称：IET IMAGE PROCESSING （ 影响因子：2.3； 五年影响因子：2.3 ）

ISSN： 1751-9659

年卷期： 2023 年 17 卷 8 期

页码：

收录情况： SCI

摘要： Most of the shape classification methods are based on a single closed contour. However, practical shapes always have complex contours, for example, a combination of multiple open contours. How to accurately identify complex shapes is an unsolved problem. In this research, a novel method is proposed to classify complex shapes. The proposed method firstly encodes a complex shape to an angle code and a sparsity code, then input these codes to a 1-D CNN for extracting features and classification. Experiments on two datasets show this novel method is superior in terms of classification accuracy. These two datasets are practical shape dataset collected by this paper on internet and MPEG-7 CE-1 Part B. The proposed method achieves higher classification accuracy than compared methods. In order to show the performance of the proposed method on each class, the accuracy on each class is analyzed. Ablation experiment is conducted to show the contribution of each module in the network. The result shows that each module is meaningful in the network, because without any module the accuracy drops.