Content-Based Image Retrieval Using Texture Structure Histogram
Hou Gang, Qinghe Feng, Xiaoxue Zhang, Kong Jun, Ming Zhang · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2013
In this paper, we proposed a novel image feature descriptor, namely texture structure histogram (TSH) for content-based image retrieval.This method using the color and edge orientation information to describe the image texture structure information.Considering the HSV color space conforms to humans' visual perception mechanism, the feature extraction is conducted in the HSV color space.This paper puts forward the non-equal interval quantization scheme that makes the expression of the image information to be more reasonable.In feature representation phase, we use the feature fusion mechanism that makes the color and shape information merge together and get a better results.The experiment results demonstrate that the proposed method more efficient and have a high retrieval performance.