摘要
本文研究了批量采集人像时,数字标识自动识别的问题。提出了描述数字结构的n-轮廓跟踪向量的概念,建立了人像背景下数字标识自动识别系统。该系统主要由图像预处理、数字目标定位及分割、数字识别等模块构成。实验表明,本系统能够对人像背景下的数字标识进行有效的识别。
This paper studies the problem of digital identification recognition in the situation of collecting personal photos in bulk.A concept of n-contour tracing vector is put forward after the research on the structure characteristics.It establishes a digital identification automatic recognition system.This system consists of image preprocessing,digital extraction and digital identification.The result showed that this system can recognize printing numbers effectively while collecting personal images.
出处
《武汉船舶职业技术学院学报》
2012年第5期36-40,共5页
Journal of Wuhan Institute of Shipbuilding Technology
关键词
数字识别
色彩空间
n-轮廓跟踪
特征提取
digital recognition
color space
n-contour tracing
feature extraction