An Image Saliency Detection Method Based on Combining Global and Local Information

被引:0
作者
Yang, Hangxu [1 ,2 ]
Gong, Yongjian [1 ]
Wang, Kai [3 ]
机构
[1] Jinhua Polytech, Coll Mech & Elect Engn, Jinhua 321017, Zhejiang, Peoples R China
[2] Key Lab Crop Harvesting Equipment Technol Zhejian, Jinhua 321017, Zhejiang, Peoples R China
[3] Nanjing Agr Univ, Coll Engn, Nanjing 210031, Jiangsu, Peoples R China
关键词
Compendex;
D O I
10.1155/2022/1849995
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
In the field of computer vision, image saliency target detection can not only improve the accuracy of image detection but also accelerate the speed of image detection. In order to solve the existing problems of the saliency target detection algorithms at present, such as inconspicuous texture details and incomplete edge contour display, this paper proposes a saliency target detection algorithm integrating multiple information. The algorithm consists of three processes: preprocessing process, multi-information extraction process, and fusion optimization process. The frequency domain features of the image are calculated, the algorithm calculates the frequency domain features of the image, introduces power law transform and feature normalization, improves the frequency domain features of the image, saves the information of the target region, and inhibits the information of the background region. On three public MSRA, SED2, and ECSSD image datasets, the proposed algorithm is compared with other classical algorithms in subjective and objective comparison experiments. Experimental results show that the proposed algorithm can not only accurately and comprehensively extract significant target regions but also retain more texture information and complete edge information while satisfying the human visual experience. All evaluation indexes are significantly better than the comparison algorithm, showing good reliability and adaptability.
引用
收藏
页数:13
相关论文
共 50 条
  • [31] Image fusion based on MPCNN and DWT in PCB failure detection
    Yuan, Yiming, 1600, Transport and Telecommunication Institute, Lomonosova street 1, Riga, LV-1019, Latvia (18):
  • [32] Research on statistical flow of the complex background based on image method
    Huanhai, Yang, 1600, International Frequency Sensor Association (173):
  • [33] An image compression method based on multiple models for the probabilities of patterns
    Department of Management Information Systems, National Chung Hsing University, Taichung, Taiwan
    不详
    Int J Imaging Syst Technol, 4 (362-368):
  • [34] Robust image registration method based on affine geometric invariant
    Li, Bin
    Ye, Hao
    Shanghai Jiaotong Daxue Xuebao/Journal of Shanghai Jiaotong University, 2012, 46 (12): : 1881 - 1884
  • [35] Blind image forensics method based on artificially blurred edges
    Hua, Bei
    Chen, Qian
    Huang, Ruwei
    Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition), 2021, 49 (02): : 127 - 132
  • [36] Visible and infrared image fusion based on multi-level method and image contrast improvement
    Peng, Yiyue
    He, Weiji
    Gu, Guohua
    Tong, Tao
    Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering, 2013, 42 (04): : 1095 - 1099
  • [37] Smile and laugh expressions detection based on local minimum key points
    Dashti, Mina Mohammadi
    Harounir, Majid
    arXiv, 2021,
  • [38] Face recognition for access control systems combining image-difference features based on a probabilistic model
    Miwa, Shotaro
    Kage, Hiroshi
    Hirai, Takashi
    Sumi, Kazuhiko
    IEEJ Transactions on Electronics, Information and Systems, 2011, 131 (12) : 2165 - 2171
  • [39] An optimizing process for improvement design based on information integration method
    Izumi, Heikan
    Sawaguchi, Manabu
    Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C, 2013, 79 (803): : 2651 - 2663
  • [40] Research and Challenges of Medical Image Landmark Detection Based on Deep Learning
    Li, Ju-Peng
    Wang, Ying-Hui
    Li, Gang
    Tien Tzu Hsueh Pao/Acta Electronica Sinica, 2022, 50 (01): : 226 - 237