rust (Hemileia vastatrix) and leaf miner (Leucoptera coffeella) in coffee crop (Coffea arabica).
Published: 4 September 2020| Version 4 | DOI: 10.17632/vfxf4trtcg.4
Contributors:
, , Description
This dataset was used in a research paper called "Artificial intelligence for detection and quantification of rust and leaf miner in coffee crop", where we propose an algorithm and a mobile application to detect Rust and Leaf Miner in coffee leaves. We manually collected and labeled the images from a farm in Brazil, totalizing 285 images of the rust and 257 of the miner. The images are in the original captured resolution of 4000x2250 pixels. We used a simple smartphone camera to capture the images, and most parts of the photos were taken in a laboratory with a white background. We noticed that some symptoms left by rust were very small and similar to the light reflections, which makes detection harder.
Files
Institutions
Faculdade de Tecnologia de Sao Paulo
Categories
Computer Vision, Object Detection, Plant Pathology, Convolutional Neural Network