作者: Yungang Zhang , Lijin Gao , Wei Gao , Jun Liu , None
关键词:
摘要: Color and shape descriptions of an image are the most widely used visual features in content-based retrieval systems. Feature vectors for color can be combined to improve performance In this paper, a novel method integrating HSV quantization curve let transform is proposed. By analyzing properties HSV(Hue, Saturation, Value) space, new dividing quantize space into 24 non-uniform bins based on soft decision introduced histogram generation. Digital employed extracting images, as it has been proved that almost optimal sparse representation objects with edges. The generated feature then weighted retrieval, using Manhattan distance metric similiarity measure. Experiments database 565 images show performs well precision adaptability.