A Comparative Study of Deep Learning Models With Handcraft Features and Non-Handcraft Features for Automatic Plant Species Identification

A Comparative Study of Deep Learning Models With Handcraft Features and Non-Handcraft Features for Automatic Plant Species Identification

Shamik Tiwari (UPES University, India)
DOI: 10.4018/IJAEIS.2020040104

Abstract

The classification of plants is one of the most important aims for botanists since plants have a significant part in the natural life cycle. In this work, a leaf-based automatic plant classification framework is investigated. The aim is to compare two different deep learning approaches named Deep Neural Network (DNN) and deep Convolutional Neural Network (CNN). In the case of deep neural network, hybrid shapes and texture features are utilized as hand-crafted features while in the case of the convolution non-handcraft, features are applied for classification. The offered frameworks are evaluated with a public leaf database. From the simulation results, it is confirmed that the deep CNN-based deep learning framework demonstrates superior classification performance than the handcraft feature based approach.
Article Preview
Top

2. Literature Review

A considerable work is proposed in the area of leaf-based plant recognition during the last decade using image processing and computer vision methods. Kadir et al. (2011) have presented a probabilistic neural network-based classification model for 60 types of foliage plants. This work has used mean, standard deviation, skewness as colour moments, shape features, grey level co-occurrence matrix-based texture features and Polar Fourier Transform. Vein features also added to increase the accuracy of the system. The result displays that the system provides mean accuracy of 93.0833%. In a similar type of another work, Kadir et al. (211) have utilized shape, texture, vein and colour features for plant classification Flavia dataset. This dataset consists of 32 types of leaves. A neural network model is designed with these features. The average accuracy of 93.75% is attained in this work.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 12: 4 Issues (2021): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 2 Issues (2012)
Volume 2: 2 Issues (2011)
Volume 1: 2 Issues (2010)
View Complete Journal Contents Listing