—In respect of the particularity of the product images, this paper studies the relevant technologies of content-based image retrieval, and proposes an improved method including image segmentation, perceptual hash, feature point matching and normalization for the product image retrieval. Using recall and precision as the performance evaluation criteria of the product image retrieval system, this paper also studies the influence of each component in the improved method on the performance of product image retrieval through experiments.
—CBIR, image segmentation, perception hash, feature point matching, normalization.
The authors are with Beijing Institute of Technology, China (e-mail: email@example.com, firstname.lastname@example.org).
Cite:Yuzhong Zhang and Chongwen Wang, "Product Images Retrieval Based on Smart Mobile Devices," International Journal of Computer Theory and Engineering vol. 8, no. 5, pp. 403-408, 2016.