Evaluation of CNN, Alexnet and GoogleNet for Fruit Recognition

• DOI: 10.11591/ijeecs.v12.i2.pp468-475
• Publication Date: 2019-01-26T15:04:23Z
• Authors: Nur Azida Muhammad; Amelina Ab Nasir; Zaidah Ibrahim; Nurbaity Sabri
• Keywords: Fruit Recognition; Googlenet; Neural Network; 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology; Alexnet; CNN

Fruit recognition is useful for automatic fruit harvesting. application can reduce or minimize human intervention during harvesting operation. However, in computer vision, very challenging because of similar shapes, colors and textures among various fruits. Illuminations changes due to weather condition also leads a task recognition. Thus, this paper tends investigate the performance basic Convolutional Neural Network (CNN), Alexnet Googlenet recognizing nine different types fruits from publicly available dataset. The experimental results indicate that all these techniques produce excellent accuracy, but CNN achieves fastest result compared with Googlenet.