This paper is a survey of research on pattern classifier. In particular, it emphasizes on the different types of pattern classifiers and their performance factors. Pattern classifiers use the algorithms of pattern recognition to classify various input classes into their respective categories. Recently many algorithms for pattern classifiers have been proposed. However no study of the normalized performance of pattern classifier has been done. Based on the implementation methods, in general there are three types of pattern classifiers: I.C. based, optical based, and software based using general-purpose computers. Advantages, disadvantages, and performance factors of each classifiers is discussed in this survey. Based on the presented facts, we can create benchmarks and normalized performance for pattern classifiers. Our future research will focus on building a pattern classifier in accordance with the normalized performance standard.