Terroir 2020 banner
IVES 9 IVES Conference Series 9 International Terroir Conferences 9 Terroir 2020 9 History and innovation of terroir 9 Detection of spider mite using artificial intelligence in digital viticulture

Detection of spider mite using artificial intelligence in digital viticulture

Abstract

Aim: Pests have a high impact on yield and grape quality in viticulture. An objective and rapid detection of pests under field conditions is needed. New sensing technologies and artificial intelligence could be used for pests detection in digital viticulture. The aim of this work was to apply computer vision and deep learning techniques for automatic detection of spider mite symptoms in grapevine under field conditions. 

Methods and Results: RGB images of grapevine canopy attacked by the spider mite (Eotetranychus carpini Oud) were manually taken in commercial vineyard (Etxano, Basque Country, Spain) under natural day light conditions. Leaf segmentation in images was performed based on computer vision techniques, isolating target leaves with spider mite visual symptoms from the vineyard canopy. HSV colour space was used to consider colour variations representing symptoms on the leaves, separating these values from those of saturation and brightness of the image. Spider mite detection was done using Convolutional Neural Networks (CNN) models with an artificially augmented dataset for the classification of leaves with this pest symptoms. An accuracy surpassing 75% was obtained using a hold-out validation.

Conclusions: 

High accuracy proves the effectiveness of the trained model in the classification of grapevine leaves. Computer vision techniques were useful to image classification on the relevant pixels. Additionally, deep learning techniques provided a robust model to find complex features of spider mite visual symptoms.

Significance and Impact of the Study: Non-invasive technology and artificial intelligence shown promising results in the automatic detection of pests in commercial vineyards.

DOI:

Publication date: March 23, 2021

Issue: Terroir 2020

Type: Video

Authors

Inés Hernández1, Salvador Gutiérrez2, Sara Ceballos1, Ignacio Barrio1, Fernando Palacios1, Ana M. Diez-Navajas3, Javier Tardáguila1*

1Televitis Research Group. University of La Rioja, 26007 Logroño, Spain 
2Department of Computer Science and Engineering, University of Cádiz, 11519 Puerto Real, Spain 
3Department of Plant Production and Protection, NEIKER-Basque Institute for Agricultural Research and Development, Basque Research and Technology Alliance (BRTA), 01192 Arkaute, Spain

Contact the author

Keywords

Deep learning, computer vision, pests, grapevine, crop protection

Tags

IVES Conference Series | Terroir 2020

Citation

Related articles…

The impact of Saccharomyces yeasts on wine varietal aroma, wine aging and wine longevity

The objective of the present work is to assess yeast effects on the development of wine varietal aroma throughout aging and on wine longevity.

Three independent experiments were carried out; two fermenting semi-synthetic musts fortified with polyphenols and aroma precursors extracted from Tempranillo (1) or Albariño (2) grapes and with synthetic precursors of polyfunctional mercaptans (PFMs), and a third in which a must, mixture of 6 different grape varieties was used. In all cases, fermentations were carried out by different Saccharomyces cerevisiae strains and one S. kudriavzevii, and the obtained wines were further submitted to anoxic accelerated aging to reproduce bottle aging. The volatile profile of the wines was analyzed using several chromatographic procedures, in order to provide a comprehensive evaluation of wine aroma. Aroma compounds analyzed included fermentation volatile metabolites, grape-derived aroma compounds including PFMs, and Strecker aldehydes (SA).

Results revealed that the effects of yeast on wine aroma throughout its self-life extend along three main axes:

1. A direct or indirect action on primary varietal aroma and on its evolution during wine
aging.

2. The direct production of SA during fermentation and/or their delayed formation by producing the required reagents (amino acids + dicarbonyls) for Strecker degradation
during anoxic aging.

3. Producing acids (leucidic, branched acids) precursors to fruity esters. More specifically, and leaving aside the infrequent de novo formation, the action of the different strains of yeast on primary varietal aroma takes four different forms:

1.- Speeding the hydrolysis of aroma precursors, which leads to early aroma formation without changing the amount of aroma formed. In the case of labile molecules, such as linalool, the enhancement of young wine aroma implies a short-living wine. 2.- Metabolizing the aroma precursor, reducing the amounts of aroma formed, which can be of advantage for negative aroma compounds, such as TDN or guaiacol; 3.- Transforming grape components into aroma precursors, increasing the amounts of aroma formed, as for ethyl cinnamate, leucidic acid or vinylphenols; 4.- Forming reactive species such as vinylphenols able to destroy varietal polyfunctional mercaptans.

Overall, it can be concluded that the yeast carrying alcoholic fermentation not only influences fermentative wine aroma but also affects to the wine varietal aroma, to its evolution during aging and to the development of oxidative off-odors

Fine-scale projections of future climate in the vineyards of southern Uruguay

In viticulture, climate change significantly impacts the plant’s development and the quality and characteristics of wines. These variations are often observed over short distances in a wine-growing region and are linked to local features (slope, soil, seasonal climate, etc.). The high spatial variability of climate caused by local factors is often of the same order or even higher than the temperature increase simulated by the different IPCC scenarios.

A phylogenomic study reveals the major dissemination routes of ‘Tempranillo Tinto’ in the Iberian Peninsula

‘Tempranillo Tinto’ is a black-berried Iberian cultivar that originated from a hybridization between cvs. ‘Benedicto’ and ‘Albillo Mayor’ [1]. Today, it is the third most widely grown wine grape cultivar worldwide with more than 200,000 hectares of vineyards mostly distributed along the Iberian Peninsula, where it is also known as ‘Cencibel’, ‘Tinta de Toro’, ‘Tinta Roriz’, and ‘Aragonez’, among other synonyms. Here, we quantified the intra-varietal genomic diversity in this cultivar through the study of 35 clones or ancient vines from seven different Iberian wine-making regions. A comparative analysis after Illumina whole-genome sequencing revealed the presence of 1,120 clonal single nucleotide variants (SNVs).

Outline for the définition of “Terroirs Viticoles application to the area of El AIjarafe (Seville, Spain)

The grapes producing and wine making regions are différent in their use of agricultural, industrial or agroindustrial means. These means are quite often very original and/or specialised; and lately are also quite competitive. Such means are being defined with increased accuracy in the delimitation and definition of its characteristics (Paneque et al., 1996 a). Human action together with other Elements and Agents involved in the vine growing production (Reyner, 1989) over these means lead to agronomic systems with important characteristics. Finally, the transformation of the vine growing production, through different technologies (Fleet, 1992), results in the creation of products with a different acceptance and economical value in the market.

Building new temperature indexes for a local understanding of grapevine physiology

Aim: Temperature corresponds to one of the main terroir factors influencing grapevine physiology, primarily evidenced by its impact on phenology. Numerous studies have aimed at expressing time with thermal indices such as growing degree days (GDD) and have thus enabled a better modelling of grapevine responses to temperature. However, some works have highlighted the need to adapt