ScabyNet, a user-friendly application for detecting common scab in potato tubers using deep learning and morphological traits.

Common scab (CS) is a major bacterial disease causing lesions on potato tubers, degrading their appearance and reducing their market value. To accurately grade scab-infected potato tubers, this study introduces "ScabyNet", an image processing approach combining color-morphology analysis with deep learning techniques. ScabyNet estimates tuber quality traits and accurately detects and quantifies CS severity levels from color images. It is presented as a standalone application with a graphical user interface comprising two main modules. One module identifies and separates tubers on images and estimates quality-related morphological features. In addition, it enables the extraction of tubers as standard tiles for the deep-learning module. The deep-learning module detects and quantifies the scab infection into five severity classes related to the relative infected area. The analysis was performed on a dataset of 7154 images of individual tiles collected from field and glasshouse experiments. Combining the two modules yields essential parameters for quality and disease inspection. The first module simplifies imaging by replacing the region proposal step of instance segmentation networks. Furthermore, the approach is an operational tool for an affordable phenotyping system that selects scab-resistant genotypes while maintaining their market standards.

Visible - Near infrared hyperspectral dataset of healthy and infected apple tree leaves images for the monitoring of apple fire blight.

This dataset consists of three groups of hyperspectral images of apple tree plants. The first group of images consists of a temporal monitoring of seven apple tree plants, infected with fire blight (Erwinia amylovora), and six control plants over a period of 15 days. The second group of images includes a temporal monitoring of three infected plants, seven plants subjected to water stress, and seven control plants. The third group of images corresponds to acquisitions made in the orchard on nine trees showing symptoms of fire blight and six control trees. The pixel locations of infected areas have been provided for all images featuring symptomatic plants.

Designing a Proximal Sensing Camera Acquisition System for Vineyard Applications: Results and Feedback on 8 Years of Experiments.

The potential of image proximal sensing for agricultural applications has been a prolific scientific subject in the recent literature. Its main appeal lies in the sensing of precise information about plant status, which is either harder or impossible to extract from lower-resolution downward-looking image sensors such as satellite or drone imagery. Yet, many theoretical and practical problems arise when dealing with proximal sensing, especially on perennial crops such as vineyards. Indeed, vineyards exhibit challenging physical obstacles and many degrees of variability in their layout. In this paper, we present the design of a mobile camera suited to vineyards and harsh experimental conditions, as well as the results and assessments of 8 years' worth of studies using that camera. These projects ranged from in-field yield estimation (berry counting) to disease detection, providing new insights on typical viticulture problems that could also be generalized to orchard crops. Different recommendations are then provided using small case studies, such as the difficulties related to framing plots with different structures or the mounting of the sensor on a moving vehicle. While results stress the obvious importance and strong benefits of a thorough experimental design, they also indicate some inescapable pitfalls, illustrating the need for more robust image analysis algorithms and better databases. We believe sharing that experience with the scientific community can only benefit the future development of these innovative approaches.

Combination of multivariate curve resolution with factorial discriminant analysis for the detection of grapevine diseases using hyperspectral imaging. A case study: flavescence dorée.

Hyperspectral imaging is an emergent technique in viticulture that can potentially detect bacterial diseases in a non-destructive manner. However, the main problem is to handle the substantial amount of information obtained from this type of data, for which reliable data analysis tools are necessary. In this work, a combination of multivariate curve resolution-alternating least squares (MCR-ALS) and factorial discriminant analysis (FDA) is proposed to detect the flavescence dorée grapevine disease from hyperspectral imaging. The main purpose of MCR-ALS in this work was to provide chemically meaningful basic spectral signatures and distribution maps of the constituents needed to describe both healthy and infected leaf images by flavescence dorée. MCR scores (distribution maps) were used as the starting information for FDA to distinguish between healthy and infected pixels/images. Such an approach is presumably more powerful than the direct use of FDA on the raw imaging data, since MCR scores are compressed and noise-filtered information on pixel properties, which makes them more suitable for discrimination analysis. High levels of correct pixel discrimination rates (CR = 85.1%) for the MCR-ALS/FDA discrimination model were obtained. The model presents a lesser ability to determine infected leaves than healthy leaves. Nevertheless, only two images were misclassified. Therefore, the proposed strategy constitutes a good approach for the detection of flavescence dorée that could be potentially used to detect other phytopathologies.

A novel robust PLS regression method inspired from boosting principles: RoBoost-PLSR.

The calibration of Partial Least Square regression (PLSR) models can be disturbed by outlying samples in the data. In these cases the models can be unstable and their predictive potential can be depreciated. To address this problem, some robust versions of the PLSR Algorithm were proposed. These algorithms rely on the downweighting of these outliers during calibration. To this end, it is necessary to estimate an inconsistency measurement between the samples and the model. However, this estimation is not trivial in high dimensions. This paper proposes a novel robust PLSR algorithm inspired from the principles of boosting: RoBoost-PLSR. This method consists of realising a series of one latent variable weighted PLSR. RoBoost-PLSR is compared with the PLSR algorithm calibrated with and without outliers and also with Partial Robust M-regression (PRM), a reference robust method. This evaluation is conducted on the basis of three simulated datasets and a real dataset. Finally Roboost-PLSR proves to be resilient to the tested outliers, and can achieve the performances of the reference PLSR calibrated without any outlier.

An annotated image dataset of downy mildew symptoms on Merlot grape variety.

This article introduces a dataset of high-resolution colour images of grapevines. It contains 99 images acquired in the vineyard from a cruising tractor. Each image includes the full foliage of a grapevine plant. These images display a diverse range of symptoms caused by the grapevine downy mildew (Plasmopara viticola), a major fungal disease. The dataset also includes various confounding factors, i.e. anomalies that are not related to the disease.  These anomalies are the natural and common phenomena affecting vineyards such as results of mechanical wounds, necroses, chemical burns or yellowing and discolorations due to nutritional or hydric deficiencies. Images were acquired in-situ on "Le Domaine de la Grande Ferrade" a public experimental facility of INRAE, in the area of Bordeaux. Acquisitions took place at early fruiting stages (BBCH 75-79) corresponding to the main sanitary pressure during growth. The acquisition device, embedded on a vine tractor, is composed of an industrial colour camera synchronised with powerful flashes. The purpose of this device is to produce a "day for night" effect that mitigates the variation of sunlight. It enables to homogenise images acquired during different weathers and time of the day and to ensure that the foreground (containing foliage) displays appropriate brightness, with minimum shadows while the background is darker. The images of the dataset were annotated manually by photo-interpretation with a careful review of expertise regarding phytopathology and physiological disorders. The annotation process consists in associating pixels with a class that defines its membership to a type of organ and its physiological state. Pixels from healthy, symptomatic or abnormal grapevine tissues were labelled into seven classes: "Limbus", "Leaf edges", "Berries", "Stems", "Foliar mildew", "Berries mildew" and "Anomalies". The annotation is achieved with the GIMP2 software as mask images where the value attributed to each pixel corresponds to one of the seven considered classes.

In Field Detection of Downy Mildew Symptoms with Proximal Colour Imaging.

This paper proposes to study the potentialities of on-board colour imaging for the in-field detection of a textbook case disease: the grapevine downy mildew. It introduces an algorithmic strategy for the detection of various forms of foliar symptoms on proximal high-resolution images. The proposed strategy is based on structure-colour representations and probabilistic models of grapevine tissues. It operates in three steps: (i) Formulating descriptors to extract the characteristic and discriminating properties of each class. They combine the Local Structure Tensors (LST) with colorimetric statistics calculated in pixel's neighbourhood. (ii) Modelling the statistical distributions of these descriptors in each class. To account for the specific nature of LSTs, the descriptors are mapped in the Log-Euclidean space. In this space, the classes of interest can be modelled with mixtures of multivariate Gaussian distributions. (iii) Assigning each pixel to one of the classes according to its suitability to their models. The decision method is based on a "seed growth segmentation" process. This step exploits statistical criteria derived from the probabilistic model. The resulting processing chain reliably detects downy mildew symptoms and estimates the area of the affected tissues. A leave-one-out cross-validation is conducted on a dataset constituted of a hundred independent images of grapevines affected only by downy mildew and/or abiotic stresses. The proposed method achieves an extensive and accurate recovery of foliar symptoms, with on average, a 83% pixel-wise precision and a 76% pixel-wise recall.