Case Studies
Real deployments, real results.
PashuTham: Wildlife Intrusion Detection and Prevention in Farmlands Using Artificial Intelligence
AiProff.ai's AI-Driven Initiative in Agricultural Security and Reforms
India is an agriculture-driven country. More than half of the Indian workforce is employed directly by the primary sector, contributing nearly 20.2% to the country's GDP. Yet, wildlife intrusion causes substantial crop damage every year, in some states like Coastal Odisha, losses reach 50,60%, sometimes 100%.
At AiProff.ai, we believe Artificial Intelligence can play a pivotal role in addressing this challenge. Under the leadership of Senior Data Scientist Nitin Saraswat, we pioneered PashuTham, an AI-driven solution for wildlife intrusion detection and deterrence.
The Agricultural Challenge: A Glimpse of Reality
Farmers nationwide have consistently reported extensive crop damage caused by wildlife, elephants, wild boars, cows, monkeys, and deer, resulting in financial losses amounting to crores over the years, alongside severe emotional tolls.
Key statistics underscore the gravity of the challenge:
- ₹1.65 Cr (~$200,000 USD), Compensation in 2022, Coimbatore Region.
- ₹32,500 / hectare, Government declared compensation for 50% crop damage.
- 50,000+ incidents of crop damage by wildlife animals annually in India.
- 341 incidents of losses reported in Pune alone between April 2022 and March 2023.
Addressing Limitations of Current Solutions
Three primary methodologies are currently employed in India, each with significant shortcomings:
- Clutch Wire with 10V Battery: Delivers electric shocks but poses safety risks to humans and children, with frequent reinstallation needs and escalating costs.
- Barbed Wire Installations: Recurring labour and material costs for seasonal installation and removal, time-consuming and labour-intensive.
- Manual Farmer Surveillance: The most inefficient and perilous method, exposing farmers to wildlife encounters and adverse weather with no scalable, long-term solution.
Wildlife intrusion is a daily reality for farmers across India
Objectives & Methodological Overview
PashuTham is built around three pivotal objectives:
- Inbound and Outbound Detection: AI model training to detect whether animals are approaching farmland or lingering on its periphery, enabling timely interventions.
- Animal Detector Alarm System: Activates alarms when an animal ventures within ~2 feet of the farmland perimeter using sophisticated audio-visual parameters.
- Monitoring & Feedback: Advanced AI/ML capabilities to monitor animal movements post-alarm, deploying additional alarms strategically to ensure the animal exits definitively.
Solution Design: Core Components
Phase 1, Initial Deployment and Monitoring:
- Tailored Camera Configuration: Specialised cameras deployed across targeted farmland areas for comprehensive coverage.
- Intelligent Threat Detection: Proprietary algorithms identify and capture relevant video clips featuring potential wildlife threats.
- PashuTham Mobile Application: Secure real-time video feeds and actionable insights for authorised farmers.
- Soil Moisture Sensor: Real-time soil water content data for informed irrigation decisions.
- Voice-Based Chatbot: AI/NLP-powered assistant for weather updates, crop recommendations, pest control, and irrigation guidance.
PashuTham device deployed at a farmland perimeter
Phase 2: Continuous Improvement
- Localised AI Enhancement: Algorithms continuously refined using EDGE computing to discern genuine threats from false positives.
- Adaptive Alert Mechanisms: Alert parameters dynamically adjusted to optimise effectiveness and mitigate deterrent habituation in wildlife populations.
Evaluation Metrics & Advantages
Success is evaluated against:
- Precision in Threat Detection: Reduction of false positives and accurate identification of wild animals and trajectories.
- Efficiency in Animal Deterrence: Ability to prompt wild animals to vacate the protected area upon alarm activation.
- Economic Viability: Cost-effectiveness of monthly maintenance ensuring sustainable deployment.
- Operational Continuity: Seamless integration with existing farmland operations.
- Scalability: Easy replication across diverse agricultural landscapes.
- Sustainability: Solar-powered operation aligned with eco-friendly practices.
- Real-Time Monitoring: Mobile application enabling farmers to monitor and control security parameters.
Insights & Future Prospects
- Data-Driven Decision-Making: Ongoing data collection and analysis to enhance system accuracy.
- Community Engagement: Local farmer involvement in shaping effective, sustainable solutions.
- Advanced AI Integration: Future iterations incorporating predictive analytics based on historical data and environmental factors.
- IoT Integration: Enhanced real-time monitoring through Internet of Things devices.
- Collaborative Partnerships: Strategic alliances with environmental agencies and research institutions for eco-friendly solutions.
Conclusion
AiProff.ai is confident that PashuTham will catalyse a transformative shift in Indian agriculture through the synergy of Artificial Intelligence and technological innovation. By staying agile, innovative, and committed to our mission, we are well-positioned to shape a more resilient and sustainable future for agriculture, where technology and ecology coexist harmoniously.
AiProff.ai excels at creating state-of-the-art AI/ML based solutions for Government, SMB, Large Enterprises and Academic Institutions. Owing to our cost-efficient and optimal approach, we are able to lower the entry barrier for organisations of all sizes for leveraging cutting-edge AI/ML solutions and expedite Time to Market.