Efficient Photo Image Retrieval System Based on Combination of Smart Sensing and Visual Descriptor

摘要

In this paper, we propose a novel efficient photo image retrieval method that automatically indexes for the searching of relevant images using a combination of geo-coded information and content-based visual features. A photo image is labeled with its GPS (Global Positioning System) coordinates at the moment of capture, and the label leads to generating a geo-spatial index with three elements of latitude, longitude and image view direction. Then, content-based visual features are extracted, and combined with the geo-spatial information for indexing and retrieving the photo images. For user's querying process, the proposed method adopts two steps as a progressive approach, filtering the relevant subset prior to using a content-based ranking function. To evaluate the performance of the proposed algorithm, we assess the simulation performance in terms of average precision and F-score using a natural photo collection. Comparing the proposed approach to search using visual feature alone, an improvement of 20.8% (61.6-40.8) was observed. The experimental results show that the proposed method exhibited a slight enhancement of around 7.2% (61.6-54.4) in retrieval effectiveness, compared to previous work. These results reveal that a combination of context and content analysis is markedly more efficient and meaningful than using only visual feature for image retrieval.