Performance Evaluation of Semantic Based Image Retrieval System from Cyberspace (SBIRSC) Framework

With the popularity and remarkable development of digital imaging technology and its usage in various domains, an extraordinary level of accuracy required in the existing image retrieval techniques. The retrieval of the requested images on the basis of the user query by convoluted features in cyberspace is a very big challenge. Various researchers were already proposed multiple techniques for image retrieval. But, after concluding the detailed research, the authors have analyzed that most of the search-engines does not produce an accurate result. In this context, a new generation of digital images retrieval from cyberspace applications must be developed. This paper presents the performance evaluation of the proposed semantic-based image retrieval system from the cyberspace (SBIRSC) framework. The proposed framework was tested on various datasets like Flickr, Shutterstock, Pixabay, Everypixel, Bing available on the cyberspace. The performance of the proposed framework is evaluated by using accuracy and response time.