A Novel Algorithm For The Multi-Level Semantic Information Of Image Searching

To make e-commerce information searching across Internet more efficient, e-commerce image searching becomes increasingly important. In this paper, A new multi-level semantic frames is first introduced. Based on grayscale histogram and Gabor texture analysis, we propose a novel method to connect the e-commerce Image to multi-level semantic frames. The method can be applied to e-commerce image information retrieval. Experiment on an image database of about 8,600 general-purpose images show that the algorithm perform well in precision and adaptability.