Comparison of Various Similarity Measures for Average Image Hash in Mobile Phone Application

IAES International Conference on Electrical Engineering, Computer Science and Informatics

电工学;计算机科学

Chaerul Haviana, Sam Farisa^1 ; Taufik, Muhammad^1

Sultan Agung Islamic University, Semarang, Central Java, Indonesia^1

关键词: Average image;    Computing capability;    Computing cost;    Content based image retrieval;    Image hash;    Mobile phone applications;    Similarity measure;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/190/1/012013/pdf
DOI  :  10.1088/1757-899X/190/1/012013

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

One of the main issue in Content Based Image Retrieval (CIBR) is similarity measures for resulting image hashes. The main key challenge is to find the most benefits distance or similarity measures for calculating the similarity in term of speed and computing costs, specially under limited computing capabilities device like mobile phone. This study we utilize twelve most common and popular distance or similarity measures technique implemented in mobile phone application, to be compared and studied. The results show that all similarity measures implemented in this study was perform equally under mobile phone application. This gives more possibilities for method combinations to be implemented for image retrieval.