Multispectral vineyard segmentation: A deep learning comparison study

Climate change and the appearance of pests and pathogens are leading to the disappearance of palm groves of Phoenix canariensis in the Canary Islands. Traditional pathology diagnostic techniques are resource-demanding and poorly reproducible, and it is necessary to develop new monitoring methodologies. This study presents a tool to identify individuals infected by Serenomyces phoenicis and Phoenicococcus marlatti using UAV-derived multispectral images and machine learning. In the first step, image segmentation and classification techniques allowed us to calculate a relative prevalence of affected leaves at an individual scale for each palm tree, so that we could finally use this information with labelled in situ data to build a probabilistic classification model to detect infected specimens. Both the pixel classification performance and the model’s fitness were evaluated using different metrics such as omission and commission errors, accuracy, precision, recall, and F1-score. It is worth noting the accuracy of more than 0.96 obtained for the pixel classification of the affected and healthy leaves, and the good detection ability of the probabilistic classification model, which reached an accuracy of 0.87 for infected palm trees. The proposed methodology is presented as an efficient tool for identifying infected palm specimens, using spectral information, reducing the need for fieldwork and facilitating phytosanitary treatment.

Section: Discussionmentioning

confidence: 99%

UAV-Based Disease Detection in Palm Groves of Phoenix canariensis Using Machine Learning and Multispectral Imagery

Casas,

Arbelo,

Moreno-Ruiz

et al. 2023

“…Correctly classifying UAV images to recognize vines, soil, and weeds/cover crops is one of the most important steps to ensure the high quality of the produced maps. For this purpose, many advanced techniques have been developed in the last few years, with Deep Learning algorithms being seemingly the most promising ones [24]. However, these techniques are still experimental and need more refining.…”

Section: Images Alignment Segmentation and Consistency Analysismentioning

confidence: 99%

From Space to Field: Combining Satellite, UAV and Agronomic Data in an Open-Source Methodology for the Validation of NDVI Maps in Precision Viticulture

Govi,

Pappalardo,

De Marchi

et al. 2024

Recent GIS technologies are shaping the direction of Precision Agriculture and Viticulture. Sentinel-2 satellites and UAVs are key resources for multi-spectral analyses of vegetation. Despite being extensively adopted in numerous applications and scenarios, the pros and cons of both platforms are still debated. Researchers have currently investigated different aspects of these sources, mainly comparing different vegetation indexes and exploring potential relationships with agronomic variables. However, due to the costs and limitations of such an approach, a standardized methodology for agronomic purposes is still missing. This study aims to fill such a methodology gap by overcoming the potential flaws or shortages of previous works. To achieve this, an image acquisition campaign covering 6 months and over 17 hectares was carried out, followed by an NDVI comparison between Sentinel-2 and UAV to eventually explore relationships with agronomic variables. Comparative analyses were performed by using both classical (Ordinary Least Squares regression and Pearson Correlation) and spatial (Moran’s Index) statistical approaches: here, 90% of cases show r and MI scores above 0.6 for plain images, with these scores expectedly lowering to 72% and 52% when considering segmented images. Moreover, NDVI thematic maps were classified into clusters and validated by the Chi-squared test. Finally, the relationship and distribution of agronomic variables within NDVI and clustered maps were consistently validated through the ANOVA test. The proposed open-source pipeline allows to strengthen existing UAV and satellite applications in Precision Agriculture by integrating more agronomic variables.

“…Geometrical information from depth maps, DEMs, LiDAR data and photogrammetric reconstructions were also assessed [11,[43][44][45], showing that this information improves the baseline performance. DL approaches for semantic segmentation and skeletonization algorithms have also been discussed [46,47]. Further insight into this field is provided by Li et al [48].…”

Section: Classification Of Hyperspectral Imaging With ML and Dlmentioning

confidence: 99%

Classification of Grapevine Varieties Using UAV Hyperspectral Imaging

López,

Ogayar,

Feito

et al. 2024

Classifying grapevine varieties is crucial in precision viticulture, as it allows for accurate estimation of vineyard row growth for different varieties and ensures authenticity in the wine industry. This task can be performed with time-consuming destructive methods, including data collection and analysis in the laboratory. In contrast, unmanned aerial vehicles (UAVs) offer a markedly more efficient and less restrictive method for gathering hyperspectral data, even though they may yield data with higher levels of noise. Therefore, the first task is the processing of these data to correct and downsample large amounts of data. In addition, the hyperspectral signatures of grape varieties are very similar. In this study, we propose the use of a convolutional neural network (CNN) to classify seventeen different varieties of red and white grape cultivars. Instead of classifying individual samples, our approach involves processing samples alongside their surrounding neighborhood for enhanced accuracy. The extraction of spatial and spectral features is addressed with (1) a spatial attention layer and (2) inception blocks. The pipeline goes from data preparation to dataset elaboration, finishing with the training phase. The fitted model is evaluated in terms of response time, accuracy and data separability and is compared with other state-of-the-art CNNs for classifying hyperspectral data. Our network was proven to be much more lightweight by using a limited number of input bands (40) and a reduced number of trainable weights (560 k parameters). Hence, it reduced training time (1 h on average) over the collected hyperspectral dataset. In contrast, other state-of-the-art research requires large networks with several million parameters that require hours to be trained. Despite this, the evaluated metrics showed much better results for our network (approximately 99% overall accuracy), in comparison with previous works barely achieving 81% OA over UAV imagery. This notable OA was similarly observed over satellite data. These results demonstrate the efficiency and robustness of our proposed method across different hyperspectral data sources.