Le projet PlantHealth a été sélectionné dans le cadre de l’appel à Post-Docs lancé conjointement par Agropolis Fondation, les labex CEMEB et NUMEV et l’Institut Convergences #DigitAg en 2016-2017.
L’une des principales difficultés rencontrées en épidémiologie des maladies végétales est le manque de données. De plus, la reconnaissance automatique des maladies des plantes dans des flux d’images crowdsourcées représente un enjeu scientifique important. L’approche originale proposée par le projet PlantHealth pour résoudre ces problèmes consiste à s’appuyer sur des solutions d’apprentissage profond et d’apprentissage proactif afin de mettre en place un programme de sciences citoyennes.
La post-doctorante sélectionnée a récemment reçu le prix de la meilleure publication de 2020 dans la revue Computers and Electronics in Agriculture pour son article "New Perspectives on Plant Disease Characterization based on Deep Learning".
Abstract :
The control of plant diseases is a major challenge to ensure global food security and sustainable agriculture. Several recent studies have proposed to improve existing procedures for early detection of plant diseases through modern automatic image recognition systems based on deep learning. In this article, we study these methods in detail, especially those based on convolutional neural networks. We first examine whether it is more relevant to fine-tune a pre-trained model on a plant identification task rather than a general object recognition task. In particular, we show, through visualization techniques, that the characteristics learned differ according to the approach adopted and that they do not necessarily focus on the part affected by the disease. Therefore, we introduce a more intuitive method that considers diseases independently of crops, and we show that it is more effective than the classic crop-disease pair approach, especially when dealing with disease involving crops that are not illustrated in the training database. This finding therefore encourages future research to rethink the current de facto paradigm of crop disease categorization.