we have published an exciting UAV hyper-spectral dataset that we hope will spur innovation in developing machine learning solutions to crop disease diagnosis, providing global impact in ensuring food production
Anyone can enter the challenge, which is focused on the accurate diagnosis of Yellow Rust Disease
The benchmark model is based on our published paper,
We are looking forward to your submissions and the top 5 winning methods will be published in Remote Sensing Special issue of “Monitoring, Early Warning, and Scientific Management of Vegetation Pests and Diseases”
To Enter, or for more information please see the following link:
Our team in Malaysia has been working to advance our project dataset by collecting more data as often as possible.
We are working to create a high quality dataset of rice disease in drone images. The more data our partners can collect the more we hope to advance our disease detection machine learning algorithms.
This process is slow and difficult because of the limitations of agriculture, having to collect the data in time with the growing crop, identifying areas of disease (which farmers are actively trying to avoid) and travelling to the rural areas to perform the collection.
By improving out system we aim to increase the rate of collection and provide a valuable service to smallholders
Professor Liangxiu Han gave an invited talk at the Second Crop Pest and Diseases (P&D) Remote Sensing Conference on 3oth August
The conference was focused on realising smart plant production by utilising remote sensing technologies in the areas of monitoring, early warning and control of pests and diseases of crops, forests and grasslands.
Topis of discourse covered current theories, methods, models, and systems, as well as research results and discussions of developmental trends.
Professor Liangxiu Han was a Keynote Speaker at CropWatch-ICP training programme funded through United Nations, Economic and Social Commission for Asia and the Pacific (ESCAP), Aerospace Information Research Institute (AIR), Chinese Academy of Sciences (CAS)
“Precision Agriculture: A Big Data Driven, AI enabled Approach to Crop Disease Diagnosis”
Professor Han participated in a 3 – Day International Workshop discussing our work on Precision Agriculture.
Organised by the National Institute of Disaster Management (NIDM) in India, the workshop covered many aspects of driving development in symbiosis with the environment, including the role of technology in agriculture.
Professors Han’s presentation, ‘Precision Agriculture: Big data driven, AI enabled Approaches to Crop disease Diagnosis and Monitoring’ was a key contribution to this discussion
While the world is dealing with the pandemic in these troubling times we would like to express our support and appreciation to those that are working to safeguard the vulnerable and our consolation to everyone effected.
Due to the disruptions in international travel we have had to delay our scheduled project demonstration and are continuing to work towards improving our solution during these troubling times.
Today we held a meeting to demonstrate the current progress of the project with representatives from numerous industry partners. The goal was to determine how successfully our solution was in solving the issue of Bacterial Leaf Blight detection in rice fields and to discuss ways of improving it to meet the needs of different end users.
The discussion went well with great interest on behalf of the industry partners and excellent feedback to further develop and refine the project.
During a visit to a rice paddy we used our mobile application to automatically survey the field. The literal field test was a great success, our application connected to the Phantom 4 Pro drone without issue, generated a flight path and uploaded it to the drone. The drone was then able to photograph the entire field without any user input. Afterwards the photos were downloaded to the mobile device before being analysed by out disease segmentation model.
We gathered some useful data about improvements to the system and the mobile application while demonstrating the ability to easily control the drone without direct user intervention.