TY - GEN
T1 - AI Based Smart Crop Monitoring for Animal Intrusions Detection
AU - Girish, D.
AU - Rohith, S.
AU - Poojitha, J.
AU - Varsha, S.
AU - Vamshi Krishna, B. V.
AU - Sreenivasulu, K. N.
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Smart Crop Guard is that incorporates Artificial Intelligence (AI) to observe and control the entrance of animals into agricultural fields. The system employs computer vision algorithms, such as deep learning models, which allow for the visual identification of different human, birds, pets, wildlife species detection through live video footage. In this work Iota technology is employed to minimize threats even in places far away from the sources. The deployment of Smart Crop Guard will lead to increased crop production and also to the harmonious relationship between agriculture and wildlife. The proposed work operates in two modes Real-time communication (RTC) and Switching Mode (SM). In SM mode AI camera is trained to recognize wildlife species through Mobile with Net-based deep learning models. The camera processes live video feeds to detect animal intrusions and triggers appropriate responses. Upon detection, the system activates alarms, sends alerts to farmers, and deploys deterrents like sound-based systems or ultrasonic devices, eliminating the need for electric fencing. The proposed method is tested for different test cases and it is observed that efficiently detects.
AB - Smart Crop Guard is that incorporates Artificial Intelligence (AI) to observe and control the entrance of animals into agricultural fields. The system employs computer vision algorithms, such as deep learning models, which allow for the visual identification of different human, birds, pets, wildlife species detection through live video footage. In this work Iota technology is employed to minimize threats even in places far away from the sources. The deployment of Smart Crop Guard will lead to increased crop production and also to the harmonious relationship between agriculture and wildlife. The proposed work operates in two modes Real-time communication (RTC) and Switching Mode (SM). In SM mode AI camera is trained to recognize wildlife species through Mobile with Net-based deep learning models. The camera processes live video feeds to detect animal intrusions and triggers appropriate responses. Upon detection, the system activates alarms, sends alerts to farmers, and deploys deterrents like sound-based systems or ultrasonic devices, eliminating the need for electric fencing. The proposed method is tested for different test cases and it is observed that efficiently detects.
UR - https://www.scopus.com/pages/publications/105010198400
UR - https://www.scopus.com/pages/publications/105010198400#tab=citedBy
U2 - 10.1109/ICKECS65700.2025.11035851
DO - 10.1109/ICKECS65700.2025.11035851
M3 - Conference contribution
AN - SCOPUS:105010198400
T3 - Proceedings of 3rd IEEE International Conference on Knowledge Engineering and Communication Systems, ICKECS 2025
BT - Proceedings of 3rd IEEE International Conference on Knowledge Engineering and Communication Systems, ICKECS 2025
A2 - Raju, G T
A2 - Manjunatha, Kumar B H
A2 - Rangaswamy, C
A2 - Bhanumathi, S
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 3rd IEEE International Conference on Knowledge Engineering and Communication Systems, ICKECS 2025
Y2 - 28 April 2025 through 29 April 2025
ER -