Develop a smart phone's base machine vision system

Abstract

The machine vision system is widely used in manufacturing factories. It is used to inspect the quality of products. This system reduces the inspection time and human error. Recently, there are various brand of machine vision systems. However, the machine vision systems are high-cost systems. There are two types of machine vision systems: smart cameras and PC bases. Generally, the cost of camera for smart camera system is higher than PC base. Especially, the high-resolution vision system is more expensive than the low-resolution. Therefore, this study proposes to develop the machine vision system from the smart phone. Since the smart phone is continuously improving, easy to reinstall (in the hardware damaged case), high resolution and lower cost than the machine vision system in the market. Thus, it is challenging to use the smart phone as an industrial machine vision system. In this study, the smart phone base machine vision system was used to detect the missing part application. The broken foil on the top of the milk carton was used to simulate the inspection process. The results found the smart phone's base machine vision system can detect the broken foil with an accuracy = 95% and an average processing time = 0.9 sec