Sistem Deteksi Kualitas Buah Jambu Air Berdasarkan Warna Kulit Menggunakan Algoritma Principal Component Analysis (Pca) dan K-Nearest Neigbor (K-NN)

Abstract

One form of artificial intelligence is the automatic detection of images. so the system can determine precisely the type of image or it can be called computer vision. Water guava fruit is a fruit that is often encountered in Indonesia, but many of the water guavas in the community are of poor quality, thus detrimental to consumers. Therefore we need a system that can detect the quality of the water guava. The Principal Component Analysis (PCA) algorithm and the k-nearest neighbor (k-NN) algorithm can be combined to do this job. PCA is an algorithm that can convert to a group of data that is initially correlated into uncorrelated data (Principal Component). The number of Principal Components generated is the same as the original data, but can be reduced to a smaller amount and is still able to represent the original data well. Meanwhile, k-NN is a method for classifying objects based on learning data that is closest to the object. The research model used in this research is a prototype, and the development tools used are UML. In making the water guava quality detection system, the MATLAB programming language is used, and the test uses the blacbox method. The result of this system is that the system is able to produce output in the form of quality classification of water guava fruit automatically.

Keywords: Computer vision, PCA, k-NN