一种新的用于集装箱字符的识别方法

主要以集装箱字符识别为例，详尽地论述了模板匹配的基本，同时，为了提高识别率，减少计算量和运行时间，满足实际应用的需要，特别是针对在实际图像中背景与字符均是彩色信息的应用来说，其二值化得到的图像可分为两种，即黑底白字和白底黑字，为了解决同一字符需要制作正反两种不同标准模板的问题，提出了一种改进的标准模板制作方法，这种模板能够自适应地匹配二值化后的正反图像，算法简单实用，理论与实验结果均表明，改进的模板匹配方法具有较高的识别率和坚韧性。

A new recognition method to container character

Character identification is an issue that computer pattern recognition often meets. Binarization and template matching method are used usually for the printed character. In this paper, it take container character recognition for example to describe essential principle of template matching in detail. At the same time, in order to improve recognition-rate, decrease compution and operation time, meet practical requirement, especially aiming at what the background and the character are all colorful in application, which are positive and negative picture after binarization, there propose a modified adaptive template matching method. It can be fit for two kinds of inverse image after binarization, what's more, method is simple and practical. The modified method has been proved and showed it has better recognition-rate and robust.