Use of Deep Learning and k-Nearest Neighbor Algorithms for Recognition of Fruit Types

Sulthan, M Burhanis
Hartiansyah, Fiqih Rahman
Hafin, Aqid Fahri

Nusantara Journal of Computers and its Applications
Vol 10, No 1 (2025): Edisi Juni 2025

fruit recognition was done in this research specifically for fruit image. The recognition of fruit in this study can be implemented to know the number of fruits that exist. Fruit image trained into several labels (fruit types) that are classified by data testing. There are several processes and methods undertaken in this research until the classification process, one of this i.e. Gaussian filter to improve the quality of fruit image recognition. Furthermore, the feature extraction process uses Gabor filter and for feature selection, PCA technic is respectively used to select some of the best features. The selected feature will be classified using deep learning and k-nearest neighbor (k-NN) method. Moreover, the results of the processes done carried out in achieving an accuracy of 95.01%.