ISI Foundation Principal Researcher André Panisson is among the winners of the crowdAI PlantVillage Disease Classification Challenge. In a world depending on edible plants just as on oxygen, with diseases remaining a major threat to the global supply of crops, the challenge aimed at developing algorithms that can accurately diagnose a disease based on an image of a plant.
Awards have been assigned during the Applied Machine Learning Days at the École Polytecnique Fédérale de Lausanne, in Switzerland. With a talk about “Lessons learned from the PlantVillage Disease Classification Challenge”, André Panisson illustrated his winning approach, based on Convolutional Neural Networks, a set of machine learning tools that are becoming more and more common in training computers to solve many problems (from image classification to speech recognition, from skin cancer diagnosis to the ability of playing traditional game Go).
CrowdAi PlantVillage Disease Classification Challenge, link: https://www.crowdai.org/challenges/plantvillage-disease-classification-challenge