Description: Handwritten digits recognition: the handwritten digits recognition is a simple image content recognition applications, because less number of patterns have been identified (only Arabic Numbers 0 to 9, 10) and Arabic numerals are few and simple strokes, etc. Handwritten digits recognition methods relative to application domains such as face recognition, Chinese character recognition, can be used in a more flexible approach, for example, the method based on rules, the method based on finite state automata, based on statistical method and the method based on neural network, etc. At present more popular method is based on the neural network method and the method based on statistics, but no matter use which kinds of method, also need through the basic image processing techniques for image preprocessing, the input information of these methods can be attained. Therefore, the starting part of this paper is to write the whole processing flow of Arabic numeration recognition, and this process can also be used to identify other patterns in the image. Of course, this process is not unique, and can be used to identify application scenarios with different preprocessing processes based on different patterns
To Search:
File list (Check if you may need any files):
Download