Recently, indoor localization technology has attracted the attention of many scholars and IT industry enterprises. The main problem that exists in the current algorithms is time latency and localization accuracy. We propose a fast image retrieval algorithm for vision-based indoor localization system. The proposed algorithm is based on PCA, Linear Regression and GIST when an image database is built. Gist feature of database image is processed as training data set. Through PCA the key information is extracted, then we fit a model by linear regression. We compare the performance of our algorithm with C-GIST algorithm. It will be also demonstrated that our proposed algorithm takes an average of 31.5 percent less time for image retrieval in coarse matching step, and the accuracy is better than C-GIST. The most prominent contribution is the results of our algorithm can be directly applied to the localization of a region of narrower width, and the latency of the algorithm have no relation to the size of the visual database.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A PCLR-GIST Algorithm for Fast Image Retrieval in Visual Indoor Localization System


    Contributors:
    Yin, Xiliang (author) / Ma, Lin (author) / Tan, Xuezhi (author)


    Publication date :

    2018-06-01


    Size :

    505153 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English





    GIST: Grid Services Conversation Language

    Gomez, Juan Miguel / Alor-Hernandez, Giner / Posada-Gomez, Ruben et al. | IEEE | 2008



    An Effective and Fast Retrieval Algorithm for Content-Based Image Retrieval

    Liu, Pengyu / Jia, Kebin / Lv, Zhuoyi | IEEE | 2008