Classification of Durian Types Using Features Extraction Gray Level Co-Occurrence Matrix (GLCM) AND K-Nearest Neighbors (KNN)

Abstract

Durian is one of the most popular fruits because it has a delicious taste and distinctive aroma. It has different shapes and types, especially from thorns and different colors and has fruit parts that are also not the same as other parts. In terms of fruit selection, care must be taken because consumers generally still find it difficult to distinguish physically identified types of Durian fruit due to limited knowledge of the types of Durian fruit and require a relatively long time and accuracy in sorting. Therefore, there is a need for a method to sort the types of Durian fruit effectively and efficiently. Namely image segmentation based on the classification of the types of Durian fruit to help consumers. The method used is Gray Level Co-Occurrence Matrices for feature extraction, while to determine the proximity between the test image and the training image using the K-Nearest Neighbor method based on texture based on the color of the Durian fruit obtained. Extraction features using the GLCM method based on angles of 0°, 45°, 90° and 135°. Then the KNN method is used for the classification of characteristic results using K = 3. In this study, 1281 data training was used and 321 data testing was used, resulting in an accuracy of 93%.

Klasifikasi Rimpang Menggunakan Metode K-Nearest Neighbor dan Ekstraksi Ciri Gray Level Co-occurrence Matrix

“…Dengan jarak yang sudah ditentukan yaitu 1 piksel, orientasi dibuat dengan empat arah sudut: 0°, 45°, 90°, dan 135° [6]. GLCM digunakan karena memiliki kekuatan dalam menangkap citra dalam resolusi yang sama ketika ada proses rotasi [13]. Metode ini memiliki fitur/properti seperti Homogeneity, Contrast, Correlation, Energy, IDM, dan Entropy.…”

Durian Species Classification Using Deep Learning Method

“…The method achieved 100% accuracy for training, while validation testing achieved 75%. In [23], Linear Discriminant Analysis, another supervised learning algorithm, was employed for the classification of the same four durian species. Features were extracted based on shape signature and Local Binary Patterns, using 240 durian images for training and testing the model on 42,337 durian samples, achieving an accuracy of around 70%.…”

“…The K-Nearest Neighbors algorithm and digital image processing based on the Gray-Level Co-occurrence Matrix (GLCM) were used in the study [23]. The researchers employed a dataset of Durian fruits called "fruit-262" uploaded by MIHAI MINUT on Kaggle, consisting of 1,600 images.…”