Open Category Submission

834 – AgriSense

AgriSense is a comprehensive smart agriculture platform that harnesses the power of IoT and Artificial Intelligence (AI) to optimize farming practices, boost productivity, and foster a connected agricultural community. Designed for both real-world impact and educational enrichment, AgriSense integrates environmental sensing, intelligent data processing, and community-driven learning.

The system uses IoT sensors to monitor soil moisture, temperature, humidity, and NPK levels, transmitting data to a cloud-based backend via ESP32 microcontrollers. A dynamic React dashboard presents real-time analytics and crop-specific recommendations powered by AI algorithms. These algorithms enable predictive analysis for irrigation scheduling, nutrient deficiencies, and yield forecasting.
A standout feature of AgriSense is its AI-based plant disease detection module, which uses computer vision to identify crop diseases from leaf images captured via mobile or connected cameras. This empowers farmers to take early action, reducing crop loss and minimizing pesticide use.
The platform also includes a task management system for farm operations and a community blog, encouraging collaboration between farmers, agricultural experts, and students. This knowledge-sharing hub supports discussion, troubleshooting, and documentation of local farming practices and innovations.
Built with a modular, scalable architecture, AgriSense ensures seamless integration of new sensors, machine learning models, and APIs. The backend ensures secure and reliable communication between the IoT layer and the web platform.
AgriSense also plays a pivotal role in education, serving as a hands-on learning environment for students in software engineering, AI, data science, and agronomy. It bridges theoretical knowledge and real-world impact, making it a powerful tool for project-based learning and environmental awareness.
By merging AI, IoT, and community learning, AgriSense advances sustainable farming while cultivating the next generation of tech-savvy agricultural innovators.

The inspiration behind AgriSense emerged from observing the challenges faced by local farmers—unpredictable weather, lack of real-time soil information, poor crop health visibility, and limited access to agricultural expertise. We realized that despite advances in technology, small-scale farmers often lack affordable, intelligent systems to make data-driven decisions. At the same time, students studying agriculture, environmental science, and software engineering rarely get opportunities to apply interdisciplinary skills to real-world problems. This dual gap—technological and educational—sparked the idea for AgriSense.
Our hypothesis is simple yet powerful: By integrating IoT sensors, AI models, and a collaborative web platform, we can enable data-informed agriculture while providing a practical, interdisciplinary learning experience for students. We believe this will succeed because our solution not only addresses technical gaps in farming but also supports active learning through real-world application.
AgriSense combines environmental sensors with AI-powered analytics for disease detection, irrigation planning, and crop health monitoring. The insights are visualized through a user-friendly dashboard, while the built-in task management system and community blog promote knowledge sharing. These features serve both farmers—who get actionable insights—and students—who gain hands-on experience with cutting-edge technologies.
In education, AgriSense is a scalable, modular teaching tool that allows learners to explore IoT hardware, data pipelines, computer vision models, and frontend/backend development. It transforms abstract classroom concepts into tangible impact, fostering problem-solving, critical thinking, and teamwork.
Ultimately, AgriSense reflects our belief that impactful technology should be inclusive, sustainable, and educational. It empowers users to not only grow crops but also grow knowledge.

AgriSense is a technically feasible and user-oriented solution, built on a carefully selected stack of technologies that support real-time data collection, intelligent decision-making, and collaborative engagement. The system utilizes IoT-based sensors such as NPK, soil moisture, temperature, and humidity sensors connected to ESP32 microcontrollers, which transmit data to the cloud using MQTT or HTTP protocols. The backend, developed using Node.js or Firebase, processes and stores this data, while a React-based frontend dashboard ensures responsive, real-time visualization of key metrics. AI models are integrated to enhance functionality—particularly for crop disease detection using image classification and crop advisory suggestions based on environmental trends.
To support the development of AgriSense, we require a set of low-cost, scalable IoT kits, including ESP32 boards, sensors, MAX485 modules for serial communication, and basic infrastructure for cloud hosting. Access to labeled datasets for training AI models and support from agricultural experts will also be critical for refining the recommendation engine. To validate market demand, we plan to engage local farmers through field surveys, pilot deployments, and demo sessions in collaboration with agricultural departments and educational institutions. We will also use our integrated blog and feedback system to collect continuous user input.
Core functionalities of AgriSense include real-time environmental monitoring, AI-powered disease detection through leaf image uploads, intelligent crop advisory, task and schedule management for farming activities, and a knowledge-sharing community blog. Ensuring a positive user experience is a top priority—achieved through mobile-first design, multilingual support, clean and intuitive interfaces, and simple data visualizations. Performance will be measured using a range of metrics, including sensor data accuracy, AI model prediction precision, user engagement (via dashboard visits and blog activity), task completion rates, and improvements in decision-making or crop yield reported during pilot testing.

AgriSense follows a modular, multi-layered architecture that ensures scalability and reliability. The system consists of four primary layers. The sensing layer involves the collection of real-time environmental data using sensors such as NPK, DHT11 (for temperature and humidity), and soil moisture sensors. These are connected to ESP32 microcontrollers, which serve as the system’s edge devices. Data is transmitted via UART/RS485 and sent to the cloud using Wi-Fi through MQTT or HTTP protocols, forming the data transmission layer. In the processing layer, this data is stored and processed on a cloud backend developed using Node.js or Firebase. AI modules perform analysis for pattern recognition, anomaly detection, and prediction. Finally, the presentation layer features a React.js-based web dashboard that visualizes real-time sensor readings, crop health diagnostics, task schedules, and user interactions via a community blog.
Key innovations include an AI-based crop disease detection system developed using a convolutional neural network trained on custom-labeled and public leaf image datasets. Users can upload images to receive instant predictions through a cloud-hosted inference API. Additionally, a rule-based crop advisory system provides recommendations based on environmental data, with plans to enhance it using reinforcement learning. The integrated community platform enables farmers, students, and experts to share insights, ask questions, and build agricultural knowledge collaboratively.
The development process spans six months: the first two months focus on hardware integration and backend setup, followed by AI model training and UI/UX development. The fifth month involves integration testing and deployment, and the sixth includes the launch of the community module and field testing. Performance will be measured using sensor accuracy, AI model precision, dashboard responsiveness, user engagement, and educational impact. Progress so far includes successful sensor integration and disease detection model prototyping, with frontend development actively underway.

AgriSense represents an innovative and creative approach to transforming traditional agriculture through the fusion of IoT, AI, and community-driven knowledge sharing. Unlike conventional systems that focus on isolated data points, AgriSense offers an integrated ecosystem that not only collects real-time environmental data but also converts it into actionable insights using machine learning and visual analytics. The idea originated from the realization that many farmers—especially in under-resourced communities—lack access to timely, accurate information about their crops and soil, often relying on guesswork or outdated practices. Our solution creatively addresses this gap by combining smart sensing with intelligent automation, ensuring that decisions are backed by real data and AI-driven recommendations.
One of the most innovative aspects of AgriSense is the AI-powered disease detection module, which allows users to upload leaf images and receive instant feedback on potential diseases. This empowers farmers to respond proactively rather than reactively, reducing crop loss and improving yield. Additionally, the community blog and task management system bring together farmers, agricultural experts, and students in one digital space—fostering collaboration, continuous learning, and local problem-solving. This blend of technical innovation and social connectivity enhances the platform’s relevance and adoption potential.
Creatively, AgriSense breaks away from one-size-fits-all dashboards by offering a responsive, multilingual, and mobile-friendly interface tailored to different user personas—farmers, students, and agronomists. It transforms complex sensor readings into easy-to-understand visuals and personalized crop tips. Moreover, its modular architecture makes it highly extensible, enabling future features like predictive irrigation, drone-based monitoring, or voice-based assistance. Together, these elements make AgriSense not just a product, but a smart farming assistant that redefines how agricultural data is utilized. This combination of technical depth and human-centered design is what makes the project both innovative and impactful.

AgriSense is designed with scalability and sustainability at its core, both in terms of technological infrastructure and long-term user impact. To ensure the system can handle increasing user demand, we employ a modular and cloud-based architecture using scalable backend technologies such as Firebase and Node.js, which can automatically adjust resources based on load. The frontend, built in React, supports dynamic rendering and API-driven updates to maintain performance as the number of users or sensor nodes grows. To avoid hardware-related bottlenecks, the system supports distributed deployment—where multiple ESP32 nodes operate independently but sync data to a shared cloud, ensuring reliable operation in large or segmented fields.
To further promote scalability, AgriSense uses industry-standard communication protocols (MQTT, HTTP, RS485), allowing easy integration of new sensors, data sources, or even third-party services like weather APIs or satellite data. Its AI modules are designed as microservices, making them independently upgradable or replaceable as models improve or as user-specific crops and regional conditions evolve.
Sustainability is addressed through the use of low-power sensors and microcontrollers, minimizing energy consumption in field deployments. Additionally, the platform encourages optimized water usage and fertilizer planning through data-driven insights—directly contributing to environmental conservation. We aim to support solar-powered operation in future deployments for even greater sustainability.
To foster long-term engagement, AgriSense includes a task management system that reminds users of critical farming activities and a collaborative blog space that keeps the community active and learning. Regular updates, multilingual support, and user feedback loops will allow us to continuously adapt the platform to evolving needs. Educational institutions can also integrate AgriSense into their curricula, ensuring ongoing relevance. Together, these strategies ensure that AgriSense remains effective, eco-friendly, and scalable well beyond its initial deployment.

AgriSense is deeply rooted in addressing social issues, particularly those related to agricultural sustainability, education, and rural development. By providing farmers, especially in underserved areas, with real-time data on soil conditions, climate, and crop health, AgriSense directly impacts agricultural productivity. It empowers farmers to make informed decisions that lead to higher yields, reduced waste, and more efficient use of resources such as water and fertilizers, improving both their livelihoods and the environment. Furthermore, the platform’s AI-powered disease detection helps farmers protect their crops early, reducing the risk of crop failure and the need for harmful pesticides, which can be costly and environmentally damaging.
AgriSense’s commitment to equity and inclusion is embedded in its design. The platform is multilingual, ensuring accessibility for farmers who may not speak English or have technical expertise. Its mobile-first approach ensures that users in rural or remote areas with limited access to computers can still benefit from its features. Additionally, the community blog fosters knowledge sharing between farmers, students, and agricultural experts, enabling collaboration and local problem-solving, especially for marginalized communities.
To measure the social impact of AgriSense, we will track key metrics such as improvements in crop yield, reductions in pesticide use, and increases in water efficiency. We will also monitor user engagement on the platform, especially from rural areas, and track educational outcomes for students involved in the project. Feedback from farmers and community members through surveys and direct interaction will help us adapt the solution to meet the evolving needs of the community. This responsiveness, combined with continuous updates and educational outreach, will ensure that AgriSense remains a valuable and inclusive tool for improving agricultural practices and supporting rural economies.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.