Performance Analysis of Eigenface Method for Detecting Organic and Non-Organic Waste Type

Abstract

Indonesia is one of the largest countries in Asia with a very dense population. According to data from The World Bank, human population indicators in Indonesia in 2019 increased by 270 milion people. This shows that population density in Indonesia is related to world problems related to waste generated from households. The household sector contributes as the top waste producer in Indonesia. Landfilling that occurs without any waste sorting, results in waste being more difficult to decompose and difficult to recycle. Therefore, to overcome this problem, it is necessary to increase public awareness about waste sorting and processing. We propose to create a device that can help sort organic and non-organic waste with Computer Vision-based Artificial Intelligence technology using the Eigenface method and the Internet of Things. Eigenface is a method that has a working principle by using XML files in performing face recognition. The result of testing in this system can run well, where the system detects organic objects the door of the chopping machine can open and if it detects nonorganic, the machine door is closed. The accuracy result for organics is 70% and for inorganic 75%. This is due to the lack of variation in the dataset and changes in the physical condition of the object.