A Survey On Sketch Based Image Retrieval Technique
Aishwarya A. Deore, Baisa Laxman Gunjal · International journal of advance research and innovative ideas in education · 2016
Recently, content based image retrieval is becoming a way of fast retrieval. Users are not satisfied with the usual information retrieval techniques. Retrieving the relevant information from keywords is very major task and this technique has been focused by researchers in many ways. Sketch is more natural and informative, that breaking down the language barrier and one of the strong media for communication in old era as well as in current scenarios. Hence currently sketch based image retrieval (SBIR) is also in focus for research purpose. Existing system descriptor SBIR is derived from two classic descriptor namely, SHoG and GF-HOG. This descriptor used to capture pixel-level features from image patches. It is very difficult task to isolate the features those are corresponds to certain edges from whole patches. Whereas, a line-based descriptor is highly adorable to tone down the impact of noisy edges but it suffers from noise problem. For effectiveness of SBIR it is mandatory to describe lines and their relationships, rather than describing image patches. For that matter some latest techniques should be discovered that attains the smart way of sketch matching with photo realistic images. Here based on review and research, histogram of line relationship (HLR) descriptor is having good hopes. A good object boundary selection algorithm and HLR and some other processes should be discovered so that work can be carried out on reducing the impact of noisy edges by selecting the shaping edges that best correspond to the object boundaries.