{ "fetched_at": "2026-10-19T07:43:12Z", "source_url": "https://www.kaggle.com/competitions/beyond-visible-spectrum-ai-for-agriculture-P1/data", "pages": [ { "name": "data-description", "content": "The hyperspectral imagery retrieval was conducted five times from the 25th, April to the 18th, May in 2018 to cover the entire growing season of winter wheat. A DJI S1000 UAV system (SZ DJI Technology Co Ltd., Gungdong, China) with a snapshot hyperspectral sensor was used for data acquisition. The model of the hyperspectral sensor was UHD 185 Firefly (UHD 185) which can obtain reflect radiation from the visible to near-infrared band between 450-950nm wavelength. The spectral resolution was 4nm. The raw data was recorded as a 1000*1000 panchromatic image and a 50*50 hyperspectral image with 125 bands. With a data fusion processed by the Cubert software, the output was a 1000*1000 image with 125bands, and the image was also mosaiced and orthorectified. In this work, all the images were obtained at an altitude of 30 meters, with a spatial resolution close to 2cm per pixel. \n\n# Dataset Overview Paper \n[A Deep Learning-Based Approach for Automated Yellow Rust Disease Detection from High-Resolution Hyperspectral UAV Images](https://www.mdpi.com/2072-4292/11/13/1554)\n![](https://raw.githubusercontent.com/robeson1010/Hyperspectral-competition/main/20220428113850.png)\n\n\n## Files\n\nThe training dataset
\nThe validation dataset\n\n##Submission Columns\n\n* `Id` - file name\n* `Category` - Rust, Health, Other\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": "# Introduction\n\nParticipants from all over the world can take part in the challenge. \nYou must be over 18, or of legal age in your jurisdiction, in order to participate.\n\n\n# Timing and duration\nThe Crop disease challenge will take place online from **1 July to 10 August, 2023**.\n\n\n# Creating and submitting a solution\n\n\n\n1. For the challenge, you will have access to the training datasets and a range of tools in the form of a Jupyter notebook (Starter Pack). No other datasets can be used for the development of your solution. However, pre-trained models are allowed.\n\n- The data for this competition is available in the Kaggle webpage ([Data](https://www.kaggle.com/competitions/beyond-visible-spectrum/data)). It can be downloaded or processed using Kaggle's computing services.\n\n- Here is a short code example for data reading and processing the data ([Starter](https://www.kaggle.com/code/robeson/starter)).\n\n2. Each participant will be allowed to upload their solutions every 12 hours (to avoid overload and unfair behaviors).\n\n3. The final submission of your solution must be made by **10 August, 2023**, and each valid submission must consist a Jupyter notebook with the implementation of their final algorithm to allow the Organizers to reproduce the predictions.\n\n# Challenge evaluation\nThe evaluation of the solutions submitted will be performed in two stages:\n\n1. Quantitative analysis: live scoring of the quality of your solution throughout the challenge, which will be visible on the leaderboard on this platform; \n\n2. Qualitative analysis of your code performed by the Organizers. The qualitative analysis is performed only for proving the validity and reproducibility of your solution.\n\nThe final score is based solely on the quantitative evaluation. However, if the solution turns out to be not reproducible, or if any concerns about its validity arise, the organizers reserve the right to contact the team leader of the corresponding team to resolve such issues. If the issues cannot be resolved, the team will not be eligible to receive prizes.\n\nFor more details on the metrics, see the Metrics section in the Data page.\n\n\n\n\n\n\n\n\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "# Overview\n\nEarly accurate diagnosis and quantification of crop diseases are crucial for precision management by allowing targeted interventions at the field level. Effective diagnosis and control measures are dependent on symptom recognition and severity assessment (in the vast majority of cases, diseases exhibit a range of visual symptoms, e.g. coloured spots or streaks). At present, these are mainly achieved by visual observation and estimation. \n\nRecent advancements in digital imaging have facilitated the development of a variety of image-based methods for plant management that are capable of performing automated image acquisition and analysis. These methods have shown great potential for automated crop disease diagnosis based on images from RGB (red, green and blue), thermal, multispectral, and hyperspectral sensors.\n\nParticularly, hyperspectral imagery (HSI), which contains more narrow spectral bands over a contiguous spectral range, provides detailed spectral–spatial information of the disease infestation and offers the potential to provide better diagnostic accuracy.\n\nIn this challenge, we will mainly focus on the accurate diagnosis of crop diseases from hyperspectral imagery (in this case, yellow rust disease). The benchmark model for this competition is based on our published paper on remote sensing [1]. \n\n[1. A Deep Learning-Based Approach for Automated Yellow Rust Disease Detection from High-Resolution Hyperspectral UAV Images](https://www.mdpi.com/2072-4292/11/13/1554)\n\n![](https://raw.githubusercontent.com/robeson1010/Agvision/main/imgs/hy.png)\n\n# Challenge keywords\nRemote sensing; Hyperspectral; Crop disease; Deep learning \n\n\n# Workshop/conference\n\nThis challenge is part of the conference on “[The Fourth conference of Monitoring, Early Warning, and Scientific Management of Vegetation Pests and Diseases]()” in Aug 2023.\n\n# Publication and Future Plans \n\nWe will write a summary paper of top 5 winning methods to be published in Remote Sensing Special issue of “Monitoring, Early Warning, and Scientific Management of Vegetation Pests and Diseases”. \nOur goal is to update the dataset with different types of diseases and continue to make it available for research use.\n\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "The evaluation metric for this competition is the categorization accuracy.\n\nThree types are set as follows: Rust, Health and Other\n\n## Submission Format\n\nSubmission files should contain two columns: `Id` and `Category`. \n\nThe file should contain a header and have the following format:\n```\nId\tCategory\n101.tif\tRust\n102.tif\tHealth\n103.tif\tOther", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "People", "content": "# Organisers\nProf. Liangxiu Han, Manchester Metropolitan University\nProf. Wenjiang Huang, Aerospace Information Research Institute, Chinese Academy of Sciences\n\n# Technical Chairs\nDr. Xin Zhang, Manchester Metropolitan University\nDr. Yingying, Dong Aerospace Information Research Institute, Chinese Academy of Sciences\nTam Sobieh, Manchester Metropolitan University\n\n# Primary contact\nLiangxiu Han - Email: l.han@mmu.ac.uk\nWenjiang Huang - Email: huangwj@radi.ac.cn", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }