Gabriel António Carneiro

Abstract
Precision Agriculture (PA) has emerged as an approach to optimize production, comprise different technology and principles focusing on how to improve agricultural production. Currently, one of the main foundations of PA is the use of artificial intelligence, through deep learning (DL) algorithms. By processing large volumes of complex data, DL enhances decision-making and boosts farming efficiency. However, these methods are hungry for annotated data, which contrasts with the scarce availability of annotated agricultural data and the costs of annotation. Self-supervised learning (SSL) has emerged as a solution to tackle the lack of annotated agricultural data. This study presents a review of the application of SSL methods to computer vision tasks in the agricultural context. The aim is to create a starting point for professionals and scientists who intend to apply these methods using agricultural data. The results of 33 studies found in the literature are discussed, highlighting their pros and cons. In most of the studies, SSL outperformed its supervised counterpart, using datasets from 4000 to 60,000 samples. Potential directions for improving future research are suggested.

Abstract
The Eurasian grapevine (Vitis vinifera L.) is one of the most extensively cultivated horticultural crop worldwide, with significant economic relevance, particularly in wine production. Accurate grapevine variety identification is essential for ensuring product authenticity, quality control, and regulatory compliance. Traditional identification methods have inherent limitations limitations; ampelography is subjective and dependent on skilled experts, while molecular analysis is costly and time-consuming. To address these challenges, recent research has focused on applying deep learning (DL) and machine learning (ML) techniques for grapevine variety identification. This study systematically analyses 37 recent studies that employed DL and ML models for this purpose. The objective is to provide a detailed analysis of classification pipelines, highlighting the strengths and limitations of each approach. Most studies use DL models trained on leaf images captured in controlled environments at distances of up to 1.2 m. However, these studies often fail to address practical challenges, such as the inclusion of a broader range of grapevine varieties, using data directly acquired in the vineyards, and the evaluation of models under adverse conditions. This review also suggests potential directions for advancing research in this field.

Abstract
Precision agriculture (PA) has advanced agricultural practices, offering new opportunities for crop management and yield optimization. The use of unmanned aerial vehicles (UAVs) in PA enables high-resolution data acquisition, which has been adopted across different agricultural sectors. However, its application for decision support in chestnut plantations remains under-represented. This study presents the initial development of a methodology for segmenting chestnut burrs from UAV-based imagery to estimate its productivity in point cloud data. Deep learning (DL) architectures, including U-Net, LinkNet, and PSPNet, were employed for chestnut burr segmentation in UAV images captured at a 30 m flight height, with YOLOv8m trained for comparison. Two datasets were used for training and to evaluate the models: one newly introduced in this study and an existing dataset. U-Net demonstrated the best performance, achieving an F1-score of 0.56 and a counting accuracy of 0.71 on the proposed dataset, using a combination of both datasets during training. The primary challenge encountered was that burrs often tend to grow in clusters, leading to unified regions in segmentation, making object detection potentially more suitable for counting. Nevertheless, the results show that DL architectures can generate masks for point cloud segmentation, supporting precise chestnut tree production estimation in future studies.

Abstract
The grapevine variety identification is important in the wine's production chain since it is related to its quality, authenticity and singularity. In this study, we addressed the data augmentation approach to identify grape varieties with images acquired in-field. We tested the static transformations, RandAugment, and Cutmix methods. Our results showed that the best result was achieved by the Static method generating 5 images per sample (F1 = 0.89), however without a significative difference if compared with RandAugment generating 2 images. The worst performance was achieved by CutMix (F1 = 0.86).

Abstract
Public lighting is crucial for maintaining the safety and well-being of communities. Current inspection methods involve examining the luminaires during the day, but this approach has drawbacks, including energy consumption, delay in detecting issues, and high costs and time investment. Utilising deep learning based automatic detection is an advanced method that can be used for identifying and locating issues in this field. This study aims to use deep learning to automatically detect burnt-out street lights, using Seville (Spain) as a case study. The study uses high-resolution night time imagery from the JL1-3B satellite to create a dataset called NLight, which is then divided into three subsets: NL1, NL2, and NT. The NL1 and NL2 datasets are used to train and evaluate YOLOv5 and YOLOv7 segmentation models for instance segmentation of streets. And then, distance outliers were detected to find the lights off. Finally, the NT dataset is used to evaluate the effectiveness of the proposed methodology. The study finds that YOLOv5 achieved a mask mAP of 57.7%, and the proposed methodology had a precision of 30.8% and a recall of 28.3%. The main goal of this work is accomplished, but there is still space for future work to improve the methodology.