期刊名称：International Journal of Advances in Engineering and Management
电子版ISSN：2395-5252
出版年度：2021
卷号：3
期号：8
页码：251-255
DOI：10.35629/5252-0308109114
语种：English
出版社：IJAEM JOURNAL
摘要：In this paper, we try to explore convolution neural network to classify animals in animal videos. Convolution neural network is a powerful machine learning tool which is trained using large collection of diverse images. In this paper, we combine convolutional neural network and SVM for classification of animals. In the first stage, frames are extracted from the animal videos. The extracted animal frames are trained using VGG19 pre-trained convolution neural network. Further, the extracted features are fed into multiclass SVM classifier for the purpose of classification. To evaluate the performance of our system, we have conducted extensive experimentation on our own dataset of 15000 images, each class containing 1000images. From the results we can easily observed that the proposed method has achieved good classification rate compared to the works in the literature.
