基于YCbCr颜色空间的快递单手写文字分割

Handwritten Text Extraction Algorithm Based on the YCbCr Color Space

目的在YCbCr颜色空间下,利用Cb颜色分量信息结合阈值分割方法,提取快递单图像手写体文字信息。方法首先将图像从RGB颜色空间转换到YCbCr颜色空间下,然后在Cb颜色分量图像下进行图像阈值分割处理操作,最后对提取出的手写体文字信息进行中值滤波去噪处理,并将该算法提取的结果与基于YCbCr颜色空间使用K均值聚类方法提取的结果在分割效果、分割时间与文字识别率上进行对比。结果利用Cb颜色分量提取出的手写体文字信息更清晰,具有更快的处理速度和更高的识别率,快递单图像平均处理时间为1.36 s,识别率为89%。结论单独利用Cb颜色分量信息提取手写文字就可得到较好的提取效果,算法简单、可行。

Objective To extract handwritten text messages from the express waybill using Cb color component information of YCbCr color space combining with threshold segmentation method. Methods Firstly, the express images were transformed from RGB color space to YCbCr color space, then image thresholding was performed under image of Cb color component. Final- ly, median filtering denoising was applied to the extracted handwritten text messages, and the segmentation result was compared with the extraction algorithm based on K-means clustering under YCbCr color space in segmentation effect, splitting time and text recognition rate. Results The results showed that the extraction of the handwritten text using Cb component tended to be clearer and with faster processing speeds and higher recognition rate than in the Lab color space, the average processing time of the express image was 1.36 s, and the identification rate was 89%. Conclusion Using Cb color component information to ex- tract the handwritten text can get better extraction effect, and the algorithm is simple and feasible.