TRACING AND RECOGNITION OF MEDICINAL HERBS IN MARUNTHUVAZH MALAI AT THE WESTERN GHATS THROUGH FEATURE EXTRACTION
Abstract
The identification and classification of the herbs using the naked eye is difficult in forest or mountain areas like Marunthuvazh Malai of Kanyakumari district. The difficulties arise because of the variations in the crops identified are inaccurate. Mostly the manual prediction is taken place in those areas which require high expertise and more human resources. In this work both plant identification and tracking system based on fuzzy empowered Hybrid artificial neural networks (FHANN) are proposed. Here the input is taken from the video signals taken by the drone camera. The input video signals are converted into images. The fuzzy logic along with the HANN is used for the classification of the specific herbs from the set of plants. Some of the herbs included in the analysis are Parsley, Dill, Oregano, Chervil, Stevia, Basil, Catnip, Fennel and Lemon Grass. This approach used artificial neural networks (ANN) in combination with the K-Nearest neighbor (KNN) as the hybrid model for the herb prediction and classification in association with the fuzzy logic. The Linear Discriminant Analysis (LDA) and Convolutional Autoencoder are used as a hybrid model for the extraction of the feature from the obtained images. This approach considers various shapes, color features, and textures specifically representing the specific herbs. The experimental results show that the proposed model provides better results in the identification and classification of the various medicinal herbs.

Authors
T. Sahila 1, A. Radhakrishnan2, V.A. Nagarajan3
University College of Engineering, Nagercoil, India

Keywords
Herbs, Artificial Neural Networks, K-nearest Neighbour, Linear Discriminant Analysis, Convolutional Auto encoder
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
000000000000
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 13 , Issue: 3 , Pages: 2960 - 2968 )
Date of Publication :
April 2023
Page Views :
301
Full Text Views :

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.