COBRA: A Content-Based Retrieval Platform for Digital Images

Philippine Information Technology Journal

Manalang, Martee Aaron L.¹  de Robles, Marie Yvette B.¹  Magno, Katrina Joy H.¹ 

关键词: content-based image retrieval;    image databases;   

DOI  :  10.3860/pitj.v4i2.2732

学科分类：计算机科学（综合）

来源: Philippine Society of Information Technology Educators

【 摘 要 】

COBRA is an online environment for content-based image retrieval algorithms. Image retrieval in COBRA involves two major steps: image features extraction and image querying. It is designed to easily add algorithms for feature extraction and matching functions that will determine the similarity of the query image and the images in the database without the need to build the whole system from scratch. The system can later be used in image processing and artificial intelligence courses where students can study and modify existing algorithms and add new algorithms for image retrieval. It does not require a high performance computing device for anyone to use the platform since computations are done on the server. To make the computation even faster, feature extraction and image querying will later be done in parallel. Currently, the system can be tested on an image database that is composed of 100,000 images of wide range. Moreover, users making queries can choose from a variety of feature extraction algorithms depending on their needs. COBRA also provides an easy-to-use graphical user interface for all types of users.

【 授权许可】

Unknown