High-performance content-based image retrieval using DFS strategy

Abstract

Image data is becoming more and more popular due to the prevalence of image capture devices. How to retrieve the images effectively and efficiently from a large number of images has been a challenging issue in recent years. To deal with such issue, the major purpose of this paper is to propose a concept- and content-aware image retrieval approach using Depth-First Search (DFS) strategy to conduct effective and efficient image semantic retrieval. For effectiveness and efficiency, since the search space is reduced into specific subspaces, the retrieval cost is decreased and the retrieval quality is increased. For semantic retrieval, our proposed method can detect the potential concepts to satisfy the user's semantic need. In the experimental result, it reveals that our proposed approach is more effective and efficient than traditional ones using Breadth-First-Search (BFS) strategy.