Plant disease detection utilizing artificial intelligence and remote sensing
Alberto Cruz: California State University, Bakersfield
<div>The rapid development of new technologies and the changing landscape of the online world (e.g., Internet of Things - IoT, cloud-based solutions) provide a unique opportunity for developing intelligent agricultural systems for precision applications. Technological advances in computer vision, mechatronics, artificial intelligence and machine learning have enabled the development and implementation of remote sensing technologies for plant symptom identification and disease management. These technologies can be used to distinguish between a variety of diseases with similar symptoms, and reduce diagnosis time and cost. Herein, a vision-based program to detect symptoms of several diseases from leaf images is presented. This system utilizes a deep learning convolutional neural network (DP-CNN) and a novel abstraction-level data fusion algorithm to improve detection accuracy. It can automatically detect plant diseases and discriminate from other disorders or pathogens, despite the strong similarity. For example, the program detects symptoms of Olive Quick Decline Syndrome (OQDS) on leaves of Olea europaea L. infected by <em>Xylella fastidiosa</em> with a true positive rate of 98.60±1.47%; symptoms of Grapevine Pierce’s disease (PD) with a 99.23±0.64% accuracy, 98.08±1.67% F1-score and 0.9761±2.05 Matthew’s correlation coefficient; and symptoms of Grapevine Yellows (GY) with 92.0% accuracy and a Matthew’s correlation coefficient of 0.832.</div>
View Presentation |
