Development of Animal-Detection System using Modified CNN Algorithm

S, Sheik Mohammed.; Sheela, T.; Muthumanickam, T.

doi:10.1109/icaiss55157.2022.10011014

Cited by 4 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sheela et.al [4], proposed work on a system that integrates a Raspberry Pi module with a PIR sensor, Thermal Imaging camera, GSM module, and hologram. The animal image that was recorded is verified using a modified CNN technique, and the user is then informed.…”

Section: Related Workmentioning

confidence: 99%

An IoT Integrated Smart Prediction of Wild Animal Intrusion in Residential Areas Using Hybrid Deep Learning with Computer Vision

G. A.,

Prabha,

Aishwarya

et al. 2024

EAI Endorsed Trans IoT

View full text Add to dashboard Cite

INTRODUCTION: The conversion of forests into human lands causes the intrusion of wild animals into the residential area. There is a necessity to prevent the intrusion of such wild animals which causes damage to properties and harm or kill humans. Human population growth leads to an increase in the exploitation of forest areas and related resources for residential and other settlement purposes. There is a need for a system to detect the entry of such animals into habitats. OBJECTIVES: This paper proposes that conversion of forests into human lands causes the intrusion of wild animals into the residential area. METHODS: Deep learning technology combined with Internet of Things (IoT) devices can be deployed in the process of restricting the entry of wild animals into residential areas. The proposed system uses deep learning techniques with the use of various algorithms like DenseNet 201, ResNet50 and You Only Look Once (YOLO). These deep-learning algorithms predict wild animals through image classification. This is done using IoT devices placed in such areas. The role of IoT devices is to transmit the computer vision images to the deep learning module, receive the output, and alarm the residents of the area. RESULTS: The main results are implementation prediction of animals for image processing Datasets used for the prediction and classification indulge the use of cloud modules. It stores the dataset for the prediction process and transfers it whenever needed. As the proposed system is a hybrid model that uses more than one algorithm, the accuracy obtained from the prediction for DenseNet 201, ResNet50 and You Only Look Once (YOLO) algorithm is 82%,92%, and 98%. CONCLUSION: The prediction of those animals is done by a deep learning model which comprises three algorithms are DenseNet 201, ResNet50 and YOLOv3. Comparing the accuracy of an algorithm with higher accuracy is considered efficient and accurate.

show abstract

Section: Related Workmentioning

confidence: 99%