A Feature Extracting Method for Chinese Character Recognition of License Plate Based on Elastic Mesh and Wavelet Transform

• DOI: 10.4028/www.scientific.net/amr.760-762.1452
• Publication Date: 2013-09-18T11:27:29Z
• Authors: Chao Zheng; Hua Yang; Xing Yang; Chao Chao Huang; Xiao Di Wu
• Keywords: Optimal distinctiveness theory; Feature (linguistics)

Low-resolution Chinese character recognition of license plate is always a difficult problem. For solving it, we must think about the distinctiveness feature and counting speed method simultaneously. In this paper, proposed simple effective extraction algorithm. First, extract statistical based on decomposing stroke with wavelet transform. Second, apply Elastic Mesh Algorithm into extracting coefficient to get structure information character. The experimental results show robust against low quality characters, such as skew, fuzzy, glue, distorted character, easy be used in actual projects advantage.