Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Ship Science and Technology》 2019-14
Add to Favorite Get Latest Update

Research on automatic image segmentation algorithms based on neural network and multi-feature

LUO Jing;XIAO Xiao-xu;Guangzhou Nanyang Polytechnic College;Guangzhou International Economics College;  
Current image segmentation algorithms have some shortcomings, such as high segmentation error rate and inadequate segmentation speed. In order to improve the accuracy and speed of image segmentation, an automatic image segmentation algorithm based on neural network and multi-features is designed. Firstly, the research progress of current image segmentation algorithms at home and abroad is summarized, and the limitations of current image segmentation are found.Then, different target features are extracted from the image, and some of the most effective feature sets are selected for image segmentation. Finally, the different regions of the image are modeled and classified by using neural network to realize image segmentation, and are computed with other image segmentation methods. The method has been tested for its superiority. The results show that the error rate of neural network and multi-feature image segmentation is low, the accuracy of image segmentation exceeds 95%, the average time of image segmentation is less than that of contrast image segmentation algorithm, and the speed of image segmentation is faster.
【Fund】: 广东省高等职业技术教育研究2018年一般课题资助项目(GDGZ18Y094);; 2018年广东省高职教育信息技术教指委教改资助项目(XXJZW2018059)
【CateGory Index】: TP183;TP391.41;U675.7
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved