Automatically Identifying Wild Animals In Camera-Trap Images With Deep Learning

International Journal of Computer Science and Engineering

© 2021 by SSRG - IJCSE Journal

Volume 8 Issue 5

Year of Publication : 2021

Authors : Kalletla Sunitha

10.14445/23488387/IJCSE-V8I5P102

How to Cite?

Kalletla Sunitha, "Automatically Identifying Wild Animals In Camera-Trap Images With Deep Learning," SSRG International Journal of Computer Science and Engineering , vol. 8,  no. 5, pp. 12-16, 2021. Crossref, https://doi.org/10.14445/23488387/IJCSE-V8I5P102

Abstract:

Human vision will comprehend and examine the pictures effectively than automatic investigation by the framework. To conquer this issue in the existing order framework, a few examinations have been done, but the output has been given uniquely for low-level picture natives. Be that as it may, the existing methodology needs the exact arrangement of pictures. AI empowers the examination of enormous amounts of information. The principle point of the AI is to cause the PCs to adapt naturally without human intercession or help and change their activities appropriately. Motion- sensor "camera traps" gather fauna pictures reasonably and furthermore regularly. In any case, the data extricated from these photos will be a costly, tedious, manual errand. In this paper, we show that such data can be consequently removed by deep learning, which is a cutting-edge kind of artificial intelligence. Convolutional neural networks(CNN) are utilized for this reason.
The model for image classification utilizing CNN is developed. This model is trained with a dataset of various creature pictures to characterize pictures. The Relu activation function is utilized for fault distinguishing proof and adjustment. The trained model is tried with various datasets of creature pictures. This model yields the creature pictures with the label of the creature's name.

Keywords:

Camera Traps, Convolutional Neural Networks(CNN), Deep learning, image classification, wildlife.

References:

[1] Shahram Taheri, Önsen Toygar, Animal classification using facial images with score-level fusion, 26 July 2018.
[2] Hung Nguyen, Sarah J. Maclagan, Tu Dinh Nguyen, Thin Nguyen, Paul Flemons, Kylie Andrews. Animal Recognition with Deep Convolutional Neural Networks for Automated Wildlife Monitoring, 18 January 2018.
[3] N Manohar, Y H Sharath Kumar, G Hemantha Kumar, Supervised and unsupervised learning in animal classification, 03 November 2016.
[4] Hayder Yousif, Jianhe Yuan, Roland Kays, Zhihai He, Fast human-animal detection from highly cluttered camera-trap images using joint background modeling and deep learning classification, 28 September 2018.
[5] Tom Mitchell, Machine Learning, Mc Graw Hill.
[6] Akshaya B, Kala M T, Convolutional Neural Network Based Image Classification And New Class Detection, Power Instrumentation Control and Computing (PICC) 2020 International Conference on, (2020) 1-6.
[7] Maissa Hamouda, Karim Saheb Ettabaa, Med Salim Bouhlel, Image and Signal Processing, vol. 10884, pp. 310, 2018.
[8] Y. Bengio, Learning deep architectures for ai, Foundations and Trends in Machine Learning, (2009) 1-127.
[9] X. Glorot and Y. Bengio, Understanding the difficulty of training deep feedforward neural networks, Proceedings of Artificial Intelligence and Statistics (AISTATS), 2010.
[10] A. Krizhevsky, I. Sutskever, and G. Hinton, Imagenet classification with deep convolutional neural networks, Neural Information Processing Systems 25, 2012.
[11] Daichi Ozaki, Hiroshi Yamamoto, Eiji Utsunomiya, Kiyohito Yoshihara, Harmful Animals Detection System Utilizing Cooperative Actuation of Multiple Sensing Devices, Information Networking (ICOIN) 2021 International Conference on, (2021) 808-813.
[12] S Madhava Prabhu, Seema Verma, A Comprehensive Survey on Implementation of Image Processing Algorithms using FPGA, Recent Advances and Innovations in Engineering (ICRAIE) 2020 5th IEEE International Conference on, (2020) 1-6.
[13] K. Simonyan and A. Zisserman, Very deep convolutional networks for large-scale image recognition, arXiv preprint, 2014.