We have developed a vision-based program to detect symptoms of Olive Quick Decline Syndrome (OQDS) on leaves of Olea europaea L. infected by Xylella fastidiosa, named X-FIDO (Xylella FastIdiosa Detector for O. europaea L.). Previous work predicted disease from leaf images with deep learning but required a vast amount of data which was obtained via crowd sourcing such as the PlantVillage project. This approach has limited applicability when samples need to be tested with traditional methods (i.e., PCR) to avoid incorrect training input or for quarantine pests which manipulation is restricted. In this paper, we demonstrate that transfer learning can be leveraged when it is not possible to collect thousands of new leaf images. Transfer learning is the re-application of an already trained deep learner to a new problem. We present a novel algorithm for fusing data at different levels of abstraction to improve performance of the system. The algorithm discovers low-level features from raw data to automatically detect veins and colors that lead to symptomatic leaves. The experiment included images of 100 healthy leaves, 99 X. fastidiosa-positive leaves and 100 X. fastidiosa-negative leaves with symptoms related to other stress factors (i.e., abiotic factors such as water stress or others diseases). The program detects OQDS with a true positive rate of 98.60 ± 1.47% in testing, showing great potential for image analysis for this disease. Results were obtained with a convolutional neural network trained with the stochastic gradient descent method, and ten trials with a 75/25 split of training and testing data. This work shows potential for massive screening of plants with reduced diagnosis time and cost.

No abstract

No abstract

The colonization behavior of the Xylella fastidiosa strain CoDiRO, the causal agent of olive quick decline syndrome (OQDS), within the xylem of Olea europaea L. is still quite controversial. As previous literature suggests, even if xylem vessel occlusions in naturally infected olive plants were observed, cell aggregation in the formation of occlusions had a minimal role. This observation left some open questions about the whole behavior of the CoDiRO strain and its actual role in OQDS pathogenesis. In order to evaluate the extent of bacterial infection in olive trees and the role of bacterial aggregates in vessel occlusions, we tested a specific fluorescence in situ hybridization (FISH) probe (KO 210) for X. fastidiosa and quantified the level of infection and vessel occlusion in both petioles and branches of naturally infected and non-infected olive trees. All symptomatic petioles showed colonization by X. fastidiosa, especially in the larger innermost vessels. In several cases, the vessels appeared completely occluded by a biofilm containing bacterial cells and extracellular matrix and the frequent colonization of adjacent vessels suggested a horizontal movement of the bacteria. Infected symptomatic trees had 21.6 ± 10.7% of petiole vessels colonized by the pathogen, indicating an irregular distribution in olive tree xylem. Thus, our observations point out the primary role of the pathogen in olive vessel occlusions. Furthermore, our findings indicate that the KO 210 FISH probe is suitable for the specific detection of X. fastidiosa.

Affective computing-the emergent field in which com-1 puters detect emotions and project appropriate expressions of their 2 own-has reached a bottleneck where algorithms are not able to 3 infer a person's emotions from natural and spontaneous facial ex-4 pressions captured in video. While the field of emotion recognition 5 has seen many advances in the past decade, a facial emotion 6 recognition approach has not yet been revealed which performs well 7 in unconstrained settings. In this paper, we propose a principled 8 method which addresses the temporal dynamics of facial emotions 9 and expressions in video with a sampling approach inspired from 10 human perceptual psychology. We test the efficacy of the method on 11 the Audio/Visual Emotion Challenge 2011 and 2012, Cohn-Kanade 12 and the MMI Facial Expression Database. The method shows an av-13 erage improvement of 9.8% over the baseline for weighted accuracy 14 on the Audio/Visual Emotion Challenge 2011 video-based frame-15 level subchallenge testing set.