Interval-valued fuzzy recognition of nameplate image sequences based on visual features
LI Zai-ming · Systems engineering and electronics · 2004
Due to the illumination, fog, rain, etc., the nameplate text images may be rotated, distored, incomplete and conglutinated, which decreases the recognition rate sharply. A new interval-valued fuzzy recognition of nameplate image sequences based on high-dimension visual features is proposed. The experimental results show that this approach is superior to several widely used approaches in the recognition of the rotated, distorted, incomplete and conglutinated text images and is very robust and promising.