摘要
提出用矩特征来解决铁路货车车号识别的问题,在车号数字特征提取的过程中将数字的形状特征与不变矩特征进行了有效结合,利用数字的凹特征、中线、圆形度等形状特征,增强了数字之间的可分性,运用矩不变性降低了车号倾斜对识别结果的影响,提高了铁路货车车号识别的识别率。实验结果证明了该方法的有效性和可行性。
Moment method was adopted to solve the problem of train license recognition.The feature of the number is a combination of shape and invariant moment.The separability of the number is improved using the feature of concave,middle line and circle measure.The invariant moment decrease the influence of the skew number on the recognition rate.The recognition rate of train license is improved.The experiments demonstrated the validity and feasibility of the method.
出处
《微计算机应用》
2010年第11期34-38,共5页
Microcomputer Applications
基金
河北省自然科学基金(08M006)资助
关键词
矩
不变矩
形状特征
特征提取
车号识别
moment
invariant moment
shape feature
feature extraction
license recognition
