Abstract: The article presents a technique for perceiving alphanumeric characters situated in the picture, in view of a formerly made data set of examples utilizing brain organizations. For this reason the convolution networks were utilized, which autonomously look for highlights that permit to recognize characters in the picture. A bigger number of convolution layers permits us to perceive a more prominent number of highlights and consequently to build the likelihood of accurately perceived characters. The primary reason for the paper is to introduce programming that perceives the alphanumeric characters in pictures and to examine the effect of the size of this data set on the program's speed and character acknowledgment proficiency. This product can likewise be utilized in more perplexing designs, like programmed interpreters or as a PC peruser. The computation of the first program that perceives single person and the second program that peruses all the text from the picture have been made in the MATLAB climate. The paper depicts the parts of this product, like the learning subsystem and the person acknowledgment subsystem. The aftereffects of the program were introduced as screen captures showing the consequences of the learning system and character acknowledgment process.

| DOI: 10.17148/IJIREEICE.2022.10416