terclim by ICS banner
IVES 9 IVES Conference Series 9 A novel dataset and deep learning object detection benchmark for grapevine pest surveillance

A novel dataset and deep learning object detection benchmark for grapevine pest surveillance

Abstract

Flavescence dorée (FD) stands out as a significant grapevine disease with severe implications for vineyards. The American grapevine leafhopper (Scaphoideus titanus) serves as the primary vector, transmitting the pathogen that causes yield losses and elevated costs linked to uprooting and replanting. Another potential vector of FD is the mosaic leafhopper, Orientus ishidae, commonly found in agroecosystems. The current monitoring approach involves periodic human identification of chromotropic traps, a labor-intensive and time-consuming process.

Therefore, there is a compelling need to develop an automatic pest detection system, leveraging the recent progress in computer vision and deep learning techniques. However, the current progress in developing such a system is hindered by the lack of effective datasets to serve as ground-truth data for the training process.

To fill this gap, our study contributes a fully annotated dataset of S. titanus and Or. ishidae from yellow sticky traps. The dataset comprises more than 400 images, with 1000 identification per class. Guided by entomologists, the annotation task involved defining bounding boxes around relevant insects with corresponding class labels.

We trained and compared the performance of state-of-the-art object detection algorithms (YOLOv8 and Faster R-CNN). Pre-processing included automatic cropping to eliminate irrelevant background information and image enhancements to improve overall quality. Additionally, we tested the impact of altering image resolution, data augmentation, and single-class detection. Preliminary results achieved a high detection accuracy, with mAP@50 and F1-score above 90%, and mAP@50-95 around 70%, allowing a first deployment as an automatic annotation support tool.

DOI:

Publication date: June 14, 2024

Issue: Open GPB 2024

Type: Poster

Authors

Giorgio Checola1*, Paolo Sonego1, Valerio Mazzoni2, Franca Ghidoni3, Alberto Gelmetti3, Pietro Franceschi1

1 Research and Innovation Centre, Digital Agriculture Unit, Fondazione Edmund Mach, S. Michele all’Adige, TN, Italy
2 Research and Innovation Centre, Plant Protection Unit, Fondazione Edmund Mach, S. Michele all’Adige, TN, Italy
3 Technology Transfer Centre, Viticulture Unit, Fondazione Edmund Mach, S. Michele all’Adige, TN, Italy

Contact the author*

Keywords

insect detection, deep learning, smart pest monitoring, flavescence dorée, insect traps

Tags

IVES Conference Series | Open GPB | Open GPB 2024

Citation

Related articles…

Relationship between terroir and vegetative potential, productivity, yield and must composition of Vitis Vinífera L. Cvs. Cabernet Sauvignon under warm climate conditions

One cultivar could produce distinct wines with typical properties and qualities different depending on its cultivated and its mesoclimatic conditions.

Post-spring frost canopy development and fruit composition in cv. Barbera grapevines

One of the effects of warming trends is the advance of budburst, increasing the frequency of spring frost-related damage. In April 2021, a severe frost event affected central and northern italian viticulture. In a cv. Barbera vineyard located in the Colli Piacentini wine district, after such occurrence, vines were tracked and growth of primary bud shoots (PBS), secondary bud shoots (SBS), and suckers (SK) was monitored, as well as their fruitfulness and fruit composition. Vine performances were then compared to those of the previous year, when no post-budburst freezing temperatures occurred. The goal of the study was to evaluate the efficacy of SBS in restoring yield loss due to PBS injuries and analyze respective contribution to fruit composition.

IMPACT OF THE WINES’ QUALITY ON THE WINE DISTILLATES’ ORGANOLEPTIC PROFILE

Brandy de Jerez (BJ) is a spirit drink made exclusively from spirits and wine distillates and is characterized by the use of casks for aging that previously contained Sherries. The quality and sensory complexity of BJ depend on the raw materials and some factors: grape variety, conditions during processing the wine and its distillation, as well as the aging in the cask. Therefore, the original compounds of the grapes from which it comes are of great interest (1 y 2) being in most cases the Airén variety. Their relationship with the quality of the musts and the wines obtained from them has been studied (3) and varies each year of harvest depending on the weather conditions (4).

PROTEOMIC STUDY OF THE USE OF MANNOPROTEINS BY OENOCOCCUS OENI TO IMPROVE MALOLACTIC FERMENTATION

Malolactic fermentation (MLF) is a desired process to decrease acidity in wine. This fermentation, carried out mostly by Oenococcus oeni, is sometimes challenging due to the wine stress factors affecting this lactic acid bacterium. Wine is a harsh environment for microbial survival due to the presence of ethanol and the low pH, and with limited nutrients that compromise O. oeni development. This may result in slow or stuck fermentations. After the alcoholic fermentation the nutrients that remain in the medium, mainly released by yeast, can be used in a beneficial way by O. oeni during MLF.

Qualitative and productive characterization of a minority variety: ‘Branco lexítimo’ in DO Ribeira Sacra (Spain)

The actual climate changes, together with the strong regulation of the European Union and Spanish government, in search of sustainable viticulture, have forced the recovery of minority varieties, expanding the range of grape varieties, as well as the possible development of wines with unique profiles. In the Ribeira Sacra DO (Spain), a comparative study of the agronomic and qualitative behavior of the ‘Branco lexítimo’ variety has been carried out, compared to the majority white variety in the DO: ‘Godello’, located in the same study plot, with identic soil and climatic conditions. The study contemplated the analysis of phenology and leaf water potential, as well as the productive results and the analysis of the must quality, during four seasons: 2018 – 2021.