Native Tree Classification With Roundlogs Using Machine Learning And CNN Models

Date
2021
Research Supervisor
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
Trees play a very important role in human life and nature. All living binges are dependent on trees in many ways. There are 1048 tree species in Bangladesh. All living beings are dependent on the tree directly or indirectly. To classify the native Bangladeshi tree, we have developed a classification system with Seven CNN models and an ensemble technique. We use two datasets to classify the data in this study. The first is a raw dataset, and the second is an augmented dataset. Vgg16, Vgg19, Resnet50, EfficientnetB3, and Densenet121 train on the raw dataset. Additionally, Resnet50, EfficientnetB7, DenseNet121, and MobileNetV2 train on the augmented dataset. Then, combine the CNN models using the ensemble learning technique to achieve greater accuracy.
Description
Keywords
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University
Printed Thesis
DOI
ISSN
ISBN