图像特征识别通过神经网络训练方法实现，是学习参考的好资料-you will need first to run the file that name "charGUI4.fig" and on the right side there is a load training set where you have to train the system first, run any data that is should be from 1 to 9 and 0 like ( 1 2 3 4 5 5 6 7 8 9 0) as many line as you want so if you have a traning data that is 10 columns by 5 lines then you have to assign that to the small boxes that name column and line. right now keep the#words as is because I plan to use it but I change my mind so that it is automatically count the word for you.

after you train the system then load any image test, from the "Load Image" once you upload the image then choose the "select" and then go to the uploaded image and select whatever you want to recognize, after that select "crop" to crop the exact part that you want to recognize after that selcet "Features Extraction" to extract the features after that choose the "recognize" to see the results up on the result box.

P.S. keep in mind that the program has been tested on the