Reading optical form from web camera with c #

In this study, the programme which can read optical form with the web camera using C#.Net platform is developed. Firstly a test mechanism is designed. A standard camera 1,3 megapixels, Philips SPC 900 NC is fixed to this mechanism ,at the same time optical forms are ensured to be set to this test mechanism in the same direction with the help of fixture. Developed programme gets the picture of optical form with the help of web camera. First the picture in the format of 640x480 pixel RGB (red, green, blue) is changed to grey level then the picture is set off by stretching contrast. According to dynamic verge rate, with Otsu Algorithm, from its six parts, the form is transformed from grey level to binary level. The interpretation of double degree picture that comes out black and white pixels is made according to the situation of pixels being black and white or the number of black and white pixels. At first, the coordinates of signs at the right top and left top side of form are ascertained. And then the central coordinates of all the options in the form are computed according to the signs. The total of black and white pixels (11x11) around the central verge belonging to the options are found. With the developed algorithm, totals of the black pixels are evaluated and registered to the database. Signed options are compared with the answer key and the number of the true, false, empty or signed wrongly are obtained. Statistics of the exam analysis is made by evaluating all the answer sheets belonging to exam. In this study as a result of the experiments done with 221 students; the process that notes each optical form down the computer base with the help of CCD camera, applying the technique of picture process and assessing all the signs in the form are almost completed in a second with various algorithm. The space in the form for coding students' number is read correctly, students' school numbers are identified rightly (%100 percent). Also the space for coding booklet type is read correctly and % 100 percent of success is gained. The average success of Test 1 and Test 2 is carried out with the percent of 99.90. In other side in 221 forms, only 7 forms are read incorrectly. 214 forms are read with zero mistake. 96.83 percent of correctness rate is obtained.