Question 5. Optical character recognition (OCR) is the electronic conversion of im- ages of typed or printed text into machine-encoded text. Figure 2 shows a section of a scanned page of text in which we wish to recognise the individual characters. The problem of the recognition of noisy characters prompted a review of some work started late in 1961 on building a mathematical FIGURE 2: Sample of some scanned text from an early paper on Optical Character Recognition. Taken from a scan of Muerle, J. L., Perceiving and Recognition Automata, Annual rept. 1 Jan-31 Dec 1962, Cornell Aeronautical Lab Inc., Buffalo NY, created by the Armed Services Technical Information Agency: http://www.dtic.mil/docs/citations/AD0296522 Using your knowledge of segmentation, feature extraction and machine- learning techniques, describe how the individual characters in Figure 2 could be extracted and then classified as belonging to particular letters of the alphabet. In addition to reasoning for your choices of technique, your answer should include pseudo-code and diagrammatic descriptions of algorithms as appropriate