Home / Current Issue / Paper 1705315
Integrated Approach for Crab Species Classification: Comparative Analysis of SVM and CNN for Accuracy Assessment
Subject area: Science,Engineering and Technology · Area of research: Machine Learning, Deep Learning
Abstract
Crab species classification is a crucial task in marine biology and ecological studies. This research presents an integrated approach using Support Vector Machines (SVM) and Convolutional Neural Networks (CNN) for the accurate classification of crab species. The study leverages SVM, a classical machine learning algorithm, and CNN, a state-of-the-art deep learning model, to explore their effectiveness in distinguishing between different crab species based on image data. The research employs pre-trained models such as MobileNetV2 and VGG16 for feature extraction and investigates their performance in predicting crab species from images. Additionally, a custom CNN model is developed and trained on a dataset comprising three crab species: Callinectes sapidus, king crab, and sally lightfoot crab. The models are evaluated and compared based on their accuracy in classifying images from a real-world crab dataset.
Keywords
Crab species classification, SVM vs CNN, Image-based identification, Marine ecology, Deep learning for biodiversity, Comparative accuracy analysis.
How to cite this paper
@article{1705315,
author = {Amit Kumar Pandey, Dr. Santosh Singh, Kalash Seetharam Shetty, Ashwani Kumar Mishra, Bipin Yadav},
title = {Integrated Approach for Crab Species Classification: Comparative Analysis of SVM and CNN for Accuracy Assessment},
journal = {Iconic Research And Engineering Journals},
year = {2023},
volume = {7},
number = {6},
pages = {228-234},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/17053151.pdf},
abstract = {Crab species classification is a crucial task in marine biology and ecological studies. This research presents an integrated approach using Support Vector Machines (SVM) and Convolutional Neural Networks (CNN) for the accurate classification of crab species. The study leverages SVM, a classical machine learning algorithm, and CNN, a state-of-the-art deep learning model, to explore their effectiveness in distinguishing between different crab species based on image data. The research employs pre-trained models such as MobileNetV2 and VGG16 for feature extraction and investigates their performance in predicting crab species from images. Additionally, a custom CNN model is developed and trained on a dataset comprising three crab species: Callinectes sapidus, king crab, and sally lightfoot crab. The models are evaluated and compared based on their accuracy in classifying images from a real-world crab dataset.},
keywords = {Crab species classification, SVM vs CNN, Image-based identification, Marine ecology, Deep learning for biodiversity, Comparative accuracy analysis.},
month = {December},
}