Citation

Shee, Jun Xian and Lim, Kian Ming and Lee, Chin Poo and Lim, Jit Yan (2023) Flower Species Recognition using DenseNet201 and Multilayer Perceptron. In: 2023 11th International Conference on Information and Communication Technology (ICoICT), 23-24 August 2023, Melaka, Malaysia.

Text 40.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Flower species recognition is the task of identifying the species of a flower from an image. It involves using computer vision techniques and machine learning algorithms to analyze the visual features of the flower in the image and match them to a known database of flower species. Flower species recognition is a challenging task due to the variations in color, shape, and size among different flower species. Accurate flower species recognition has important applications in fields such as agriculture, botany, and environmental conservation. In view of this, this research paper presents a deep learning approach for flower species recognition using a combination of DenseNet201 and MLP. The proposed model leverages the strengths of both models for enhanced performance in recognizing flower species. DenseNet201 is known for its ability to capture complex features in images, while MLP is a powerful tool for learning nonlinear relationships between features. The model achieves impressive classification results on multiple datasets, including 94.47% accuracy on Kaggle, 98.23% and 97.35% on Oxford17 for two different protocols, and 79.13% on Oxford102.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	DenseNet, Flower species recognition, Multilayer perceptron.
Subjects:	Q Science > QK Botany
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	31 Oct 2023 08:13
Last Modified:	31 Oct 2023 08:13
URII:	http://shdl.mmu.edu.my/id/eprint/11798

Downloads

Edit (login required)