It is a very simple project for someone who knows matlab and pattern recognition. The above folder contains 200 .bmp files. From these files, every 20 shows the same word and as a result 10 words are defined in the whole set (10 words, each word with 20 .bmp). Each word (20 .bmp) is one class. The input of the program should be one of these 200 words/files (.bmp) and after classification (of course training before all these) the program will say to which class the word belong to.
12 freelanceria on tarjonnut keskimäärin %project_bid_stats_avg_sub_26% %project_currencyDetails_sign_sub_27% tähän työhön
The needed work is almost done! I have worked on an equivalent task before and I did it perfectly. I will give you the code of of learning process and the final learned Neural Net.