Problem Definition, Analysis and the Solution
Rice production, also known as paddy production, is an important industry in Sri Lanka, providing a staple food for over 21.8 million people and a living for over 1.8 million farmers. Unfortunately, Sri Lankan farmers are facing significant yield challenges, including lower productivity and poor crop quality. Climate change, soil water stress, and plant health are all contributing factors to these problems. Low productivity and crop yields have a significant economic impact and can result in significant losses for farmers, jeopardizing their livelihoods. The root cause of these problems is farmers' lack of understanding of how to manage water stress in soil and plants, as well as how climate change affects plant health affect yield production.
Traditionally, hiring experienced farmers to gain insight and make recommendations for improvement has been the solution to these problems. However, this approach is expensive in terms of both human labor and time, and it may not produce the best results. Implementing autonomous Drone systems that can operate in the field indefinitely and reduce the cost of human labor is an alternative solution. Farmers can obtain valuable information about water stress and plant health, as well as real-time feedback on the impact of climate change on their crops, using these systems. This method is expected to be far more efficient and effective than traditional methods, resulting in increased productivity and crop quality for Sri Lankan farmers.
Paddy cultivation is an important industry in Sri Lanka, providing food and a living for millions of people. Farmers, on the other hand, face significant challenges in managing crop health and nutrition, particularly in terms of leaf area index (LAI), leaf nitrogen concentration, normalized difference vegetation index (NDI), and excess of green. These parameters are important indicators of paddy health and can have a big impact on crop yield and quality.
The leaf area index (LAI) compares the amount of leaf surface area to the amount of ground surface area. Because it affects the plant's ability to photosynthesize and produce energy for growth, LAI is an important parameter for paddy health.
The amount of nitrogen in a leaf is measured by its nitrogen concentration. Nitrogen is a necessary nutrient for plant growth and an important factor in crop yield and quality.
Normalized disparity The vegetation index (NDI) is a measurement of how much photosynthetically active radiation a plant absorbs. The NDI is an important parameter for paddy health because it indicates the plant's ability to absorb light and generate energy.
The amount of chlorophyll in a plant is measured by the excess of green. Chlorophyll is required for photosynthesis and is used to assess plant health.
All of these variables are important indicators of paddy health and can have a significant impact on crop yield and quality. Farmers can make informed decisions about how to manage their crops, adjust fertilizer application rates or irrigation schedules to optimize crop health and yield, and adapt to changing environmental conditions by monitoring these parameters in real-time. Using computer vision and RGB cameras to monitor these parameters provides a low-cost and accessible solution for small-scale farmers in Sri Lanka, resulting in higher profits and sustainable paddy production.
Farmers have traditionally relied on costly technologies such as satellite imagery, multispectral cameras, and LAI canopy analyzers to monitor these parameters. However, these methods are frequently prohibitively expensive for many farmers, limiting their ability to effectively manage crop health.
To address this issue, we propose a new, low-cost solution that uses computer vision and RGB cameras to continuously monitor these critical parameters. Farmers can obtain real-time data on LAI, leaf nitrogen concentration, NDI, and excess of green by using computer vision algorithms to analyze images of paddy fields, providing them with critical insights into crop health. In Addition We are getting RGB images from the Drone Camera and constructing Othomosic image and Pointcloud Visulization. We are doing Research more on Segementing the point cloud and give the farmers to better understanding and Visualization of their Paddy field with Location of where the crops are damaged and les growth rate.
This method has several advantages over traditional methods, especially in terms of cost and accessibility. Farmers can monitor crop health using existing RGB cameras at a much lower cost than satellite imagery, multispectral cameras, or LAI canopy analyzers. This method is especially advantageous for small-scale farmers who may not have access to expensive technologies.
The use of computer vision and RGB cameras also provides farmers with a one-of-a-kind tool for increasing production. Farmers can make informed decisions about crop management by obtaining real-time data on LAI, leaf nitrogen concentration, NDI, and excess of green. They can, for example, modify fertilizer application rates or irrigation schedules to improve crop health and yield. As a result, the quality of their yield may improve, resulting in higher profits.
The application of computer vision and RGB cameras is also a novel approach to the problem of managing crop health in changing environmental conditions. Farmers face significant challenges as a result of climate change, and changing weather patterns can have a significant impact on crop health. Farmers can obtain real-time data on the impact of changing weather patterns on their crops by using computer vision and RGB cameras to monitor LAI, leaf nitrogen concentration, NDI, excess of green and Point Cloud. As a result, they will be able to make more informed decisions about how to adapt their farming practices to changing conditions.
Finally, our autonomous drone monitoring solution provides a cost-effective and one-of-a-kind approach to paddy production that can significantly improve crop health, yield, and quality. Our solution Agrarian, which makes use of computer vision and RGB cameras, provides real-time monitoring of critical parameters such as LAI, leaf nitrogen concentration, NDI, and excess green, which are critical for optimizing paddy health and production. Our solution, by reducing the need for human labor, provides a more efficient and accessible method of monitoring and managing paddy production for small-scale farmers in Sri Lanka. Adoption of this self-driving drone monitoring solution can give farmers a competitive advantage and increase profits while also promoting sustainable and responsible agricultural practices.Our solution is scalable and adaptable to changing environmental conditions, making it a critical innovation for the future of paddy production in Sri Lanka.
Our solution to the problems faced by farmers in Sri Lanka is novel and has the potential to significantly increase crop yield and quality. We intend to provide farmers with real-time information about the health of their crops using drone RGB cameras and mobile applications, allowing them to make informed decisions about crop management. Drone camera parameters such as leaf health indicators, soil moisture, and crop growth patterns can be used to identify areas of concern and provide targeted solutions to improve crop yield and quality.
Furthermore, we propose matching farmers with the right mentors who can advise them on crop management practices such as crop rotation, pest control, and fertilization. Farmers can benefit from the mentorship network by making informed decisions that are critical to their livelihoods.
The combination of drone technology, mobile applications, and mentorship networks has the potential to significantly transform Sri Lanka's agricultural industry. we are empowering farmers to make informed crop management decisions by providing them with real-time information about the health of their crops. This, in turn, can lead to increased crop yield and quality, improving their livelihoods.
Our solution is a comprehensive and innovative approach to meeting the needs of Sri Lankan farmers. We are developing a sustainable and responsible solution that will have a long-term impact on the agricultural industry in Sri Lanka by leveraging modern technology and mentorship network
There are several methods for determining whether our solution is innovative:
Uniqueness: Your solution can be considered innovative if it differs from existing solutions. In this case, our approach of using drone technology and mobile applications to monitor plant health is novel and has not been widely adopted in the agricultural industry with RGB Camera.
Effectiveness: our solution is considered innovative if it outperforms existing solutions. our approach of using drone technology and mobile applications to monitor plant health can provide farmers with real-time data that can help them make informed decisions, resulting in better crop management practices and higher crop yield.
Cost-effectiveness: An innovative solution is one that provides users with cost-effective benefits. When compared to other methods of monitoring plant health, our solution reduces the need for human labor and provides farmers with real-time information at a lower cost.
Scalability: A truly innovative solution is capable of being scaled up and replicated. Our solution has the potential to be scaled up to benefit a large number of farmers across the country and replicated in other countries facing similar agricultural challenges.
cost-effectiveness: Our solution is significantly less expensive than the others because it does not necessitate the purchase and maintenance of expensive equipment such as satellites or specialized cameras. To use satellite imagery, we need a subscription of $3000 USD for daily updates, a multispectral camera that costs more than $3000 USD but can provide multiple band images, and a LAI canopy analyzer that costs $300 USD. The use of autonomous drone technology eliminates the need for human labor, lowering costs even further.
Monitoring in real time: Our solution provides farmers with real-time information about plant health conditions, allowing them to make informed decisions quickly. Satellite imagery and multi-spectral camera solutions, on the other hand, provide only periodic snapshots of crop health and can be subject to delays.
Our solution is highly portable, as farmers can easily transport a drone and a mobile application wherever they go. Farmers can now monitor their crops over large areas without the need for specialized equipment or stationary infrastructure.
Accessibility: Our solution is extremely user-friendly, as it is suitable for farmers of all skill levels. Other solutions, such as LAI canopy analyzers, on the other hand, require specialized training and knowledge to operate effectively.
In summary, our solution stands out from the crowd due to its unique combination of cost-effectiveness, real-time monitoring, portability, Visulize and accessibility. The use of autonomous drone technology and mobile applications provides farmers with a powerful tool for improving crop management practices and increasing crop yield.
This refers to the use of algorithms to analyze digital images and extract information from them. In our solution, image processing is used to analyze the RGB camera images captured by the drone and extract parameters such as LAI, leaf nitrogen concentration, NDI, and excess of green.Utilized diverse image processing techniques such as thresholding, image overlaying, and color space conversion from RGB to LAB.
These are Python libraries used for scientific computing and matrix manipulation. In our solution, these libraries are used to process the data extracted from the drone images and calculate the various plant health parameters.
These are image processing techniques used to separate the image into different regions and thresholding is used to convert the image into a binary image. In our solution, these techniques are used to separate the plant from the background and calculate the plant health parameters. dditionally, employed point cloud processing methods integrating classical machine learning and DBSCAN clustering algorithms wich will cluster planst point clouds based on Density Level.
Leveraged AWS services, including EC2, RDS, S3, and Docker, to build a robust project infrastructure.
Build Docker Image for Image Proccessing
docker build -t pointcloud-app .To Install Open3d you need a python version between 3.6 -3.10.
Install the packages that used here
pip install -r requirements.txtIn order to run the docker image
docker run -p 5000:5000 pointcloud-app