Sorting Device Coding Print Quality Machine on Packing Box Prototype Utilizing Optical Character Recognition

Abstract

The design of a system that can recognize the characters printed by the coding machine so that it can then determine the assessment of very good and bad prints, is the goal of this research. This system is useful as an indicator of an indication of a problem with the coding machine which is indicated by the number of products rejected by the system. In addition, this system is an effort to reduce printing errors so that they are sorted and do not pass to the market, thus causing misinformation to the public. This system is in the form of a moving conveyor table that has a firing area box and a reject rod as a product separator. The OCR method is programmed in a computer that is part of the system, to be able to recognize the characters printed by this coding machine. USB to TTL which functions as serial communication from the Laptop to the Microcontroller, is a complement to this tool. Testing as many as 60 times on 12 types of characters, this system has been carried out on this tool. There are six types of good characters and six types of bad characters, which have been provided. Of the 60 tests, the results were four failed tests. That is, the percentage of success of this tool is 93.33%, and the percentage of error is 6.66%.