- 期刊
- OpenAccess
Query by Sketch-an Intensified Advancement in Image Fetching Using Abbreviated Images (Sketches)
並列摘要
In computer vision techniques and applications, CBIR (Content Based Image Retrieval) is used for image retrieval. In this study we present a method SBIR (Sketch Based Image Retrieval) implemented using sketch as query for retrieving general images and facial images, used in dialectic image finding. This method is based on content based image retrieval. Edge histogram descriptor, Histogram of oriented Gradients, SIFT, Surf are used for indexing the images. These descriptors are such that they can index both input and output images. Binary outlines are given as input and a set of similar shaped images are retrieved. After indexing the comparison between the input and output images is done by distance based search. The most similar images are retrieved and displayed according to their similarity. For better results reranking can also be done. Thus the system can filter database images using these descriptors and provide refined results.