2. The OCR software breaks the image into a grid to analyse it better. It then looks through its large database of letter structures to match the segments in the grid.
3. Once it has found a likely match, it tells the user what it thinks the letter is.
4. 1. Once again, the OCR software receives text as an image
5.
6.
7. 2. The software analyses the letter for special features like curves and horizontal or vertical lines. ‘T’ is easily recognized because of the intersection of a horizontal and vertical line; ‘A’ has the triangle in the middle, and so on.
4. Once it has a match, it gives the letter to the user.
3. It compares this with all the knowledge that it has been programmed with about letter construction, and arrives at a conclusion about the letter.