Identification of Banana Ripeness using Convolutional Neural Network Approaches

Digital images have been widely used for computer vision systems of ripe bananas. Consumers or industrial companies need computer vision-based applications to determine the ripeness of bananas. The computer vision system is challenging because it processes images and must find a good method. In Sara...

Full description

Saved in:
Bibliographic Details
Published in2023 International Conference on Computer, Control, Informatics and its Applications (IC3INA) pp. 330 - 335
Main Authors Nafi'Iyah, Nur, Wardhani, Retno, Prakasa, Esa
Format Conference Proceeding
LanguageEnglish
Published IEEE 04.10.2023
Subjects
Architecture CNN
Augmentation
Companies
Computer architecture
Computer vision
Convolution
Convolution neural network
Digital control
Digital images
Identification
Neural networks
Ripe bananas
Online AccessGet full text

Cover

More Information
Summary:Digital images have been widely used for computer vision systems of ripe bananas. Consumers or industrial companies need computer vision-based applications to determine the ripeness of bananas. The computer vision system is challenging because it processes images and must find a good method. In Saranya's 2022 research, the Convolution Neural Network method is accurate. We propose to improve the CNN architecture to identify banana ripeness. We use the dataset from the Saranya 2022 research, and we augment the dataset to make it large. Proposed CNN architecture: We have 97.95% accuracy with the adam optimizer. Our proposed CNN architecture results are better than Saranya's 2022 research.
DOI:10.1109/IC3INA60834.2023.10285749