Realtime Storefront Logo Recognition

Storefront brand image (logo) recognition has many useful applications such as location recognition and advertisement. Various methods have been proposed using local and global feature matching. However, it continues to be a challenging area; unlike nature scene images, which are rich in texture details, brand images often lack texture variation, and therefore provide fewer feature points for matching. In addition, reference and test images may be acquired with different rotation, resolution, size, quality, and illumination parameters. In this paper, we propose a highly efficient and accurate approach for recognizing logos in storefront images. We separate our algorithm into two primary phases representation and recognition, each of which deals with a separate portion of the logo recognition classification algorithm. After training our classifier with 257 pre-segmented storefront logos, we were able to get an 86.0% classification accuracy on 100 test images taken from a wide variety