Irregularities that occur in the waste buildup causes the handling of waste that's not well controlled. The office workers tend to be busy which causes them to be careless about waste they produce every day. The solution to the problem can be done by recycling the waste and categorize the waste by its type. Therefore a system was made in a form of the Automatic Office Trash Bin which is able to categorize the three types of offices waste. The main component that is used in this research are Raspberry Pi, camera and infrared proximity. The process of this system is assisted by Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM). The result of the system will produce an output which are determination the type of the waste and measured based on the full amount of waste in the trash. . The average examination result of the accuracy in detecting plastic bottles and papers reach 90.63%. Cans got the best result of this examnination, it has reach 96,88%. The average examination result of the computing time reach 3,45 s for plastic bottles, 3,496 s for cans and 2,529 s for paper. The examination of the infrared proximity censor in detecting the amount of waste in the trash can reach 100%.
Copyrights © 2020