<![CDATA[Sorting product system based on dimensional specifications takes a long time and the accuracy is low if done manually. The fatigue factor of production operators results in lowering the quality of work so that the production output becomes of low quality. To overcome this, it is necessary to make sorting equipment that can distinguish the types of products by using inspection camera monitoring which is controlled by Arduino Uno. In this study, a conveyor-shaped prototype was designed which is equipped with a Webcam X6 camera and several proximity sensors to provide input to the Arduino Uno. The camera function detects the product dimensions using the Deep Learning Model SSD-MobileNet v2 method using the Python programming language and is assisted by the TensorFlow and OpenCV libraries. The results of product detection will be processed into input signals to control the Arduino Uno followed by controlling the pneumatic solenoid valves on each product. The result of reading by software is to detect circular objects by 92%, detect triangular objects, which is about 84%, and read square objects, which is about 87%.]]>
Copyrights © 2024
