School of Information Science and Engineering, Central South University, Changsha, Hunan, 410083, China

---

Abstract—Scanned certifications are widely used in China as proof of past achievements. To avoid repetitive usage of the same certification in various awards or funding applications, we design and implement a scanned-certification image retrieval system based on local object and block matching. The seal is used as a salient feature and a modified Hough round detection method is applied for extraction. And local round seals are combined with blocks attributes of the seals to build image indexes. Experimental results will demonstrate the effectiveness of the system. Additionally, we conduct experiments on single color histogram, which further demonstrates the effectiveness of the proposed image retrieval method.

Index Terms—Scanned-certification, Image retrieval, Local object, Block matching, Non-uniform histogram.

[PDF]

Cite: Hao Zhang, Ling Huang, Xianru Liu, "Effective Scanned-Certification Image Retrieval Based on Local Object and Block Matching," Journal of Software vol. 9, no. 3, pp. 648-654, 2014.