Abstract The German Research Foundation has established the priority program SPP 100+. Its subject is monitoring bridge structures in civil engineering. data‐driven methods cluster deals with use of measurements and their special global local analysis methods, which complement each other an overall multi‐scale concept order to realize condition monitoring. presented aim for damage detection, localization, quantification monitored structure. Static dynamic investigations based on mechanical...

In tropical/subtropical regions, the favorable climate associated with use of agricultural technologies, such as no tillage, minimum cultivation, irrigation, early varieties, desiccants, flowering inducing, and crop rotation, makes agriculture highly dynamic. this letter, we present Campo Verde database. The purpose creating sharing these data is to foster advancement remote sensing technology in areas tropical agriculture, primarily development testing methods for recognition mapping. a...

Abstract In this work we propose a workflow to deal with overlaid images—images superimposed text and company logos—, which is very common in underwater monitoring videos surveillance camera footage. It demonstrated that it possible use Explaining Artificial Intelligence improve deep learning models performance for image classification tasks general. A model trained classify metal surface defect, previously had low performance, then evaluated Layer-wise relevance propagation—an technique—to...

The presence of weeds in agricultural crops has been one the problems greatest interest recent years as they consume natural resources and negatively affect process. For this purpose, a model implemented to segment weed aerial images. proposed relies on DeepLabv3 architecture trained upon patches extracted from high-resolution imagery. dataset employed consisted 5 images that describes sugar beet field Germany. SegNet U-Net architectures were selected for comparison purposes. Our results...

Land Use/Land Cover (LULC) monitoring is essential for understanding Earth's surface dynamics, particularly in assessing the impact of vegetation and land changes on hydrological systems, vulnerability to extreme climatic events, forest health. In recent years, increasing dieback caused by climate change, pests, diseases has raised global concerns about ecosystem stability biodiversity. The rapid spread tree mortality need accurately capture its temporal evolution highlight necessity precise...

Abstract. The monitoring of agricultural activities at a regular basis is crucial to assure that the food production meets world population demands, which increasing yearly. Such information can be derived from remote sensing data. In spite topic’s relevance, not enough efforts have been invested exploit modern pattern recognition and machine learning methods for land-cover mapping multi-temporal, multi-sensor earth observation Furthermore, only small proportion works published on this topic...

Remote Sensing (RS) data have been increasingly applied to assess agricultural yield, production and crop condition. In tropical areas, dynamics are complex due multiple practices such as irrigation, non-tillage, rotation harvest per year. Spatial temporal information can improve the performance in land-cover type classification tasks. this context Deep Learning (DL) emerged a powerful state-of-the-art technique RS community. This work presents comparative analysis of traditional DL...

This letter evaluates metaheuristics for the supervised parameter tuning of multiresolution-region-growing segmentation. Three groups are tested in terms convergence speed and solution quality. Generalized pattern search, mesh adaptive direct Nelder-Mead represent single-solution group. Differential evolution (DE) represents population DE followed by each aforementioned hybrid metaheuristic reveals that optimization objective functions typically have countless local minima, many them leading...

The Segmentation Parameter Tuner (SPT) is a tool designed for automatic tuning of segmentation parameters. In SPT, the goodness set parameter values given by level agreement between result and reference (representing desired outcome) quantified metric selected user (empirical discrepancy methods). This used as fitness function an optimization algorithm that searches space minimum value, which expected to correspond outcome most similar reference. SPT 3.1 offers many interesting features such...

Abstract In the High Modernism period, from around 1914 to 1970, many system halls in steel construction were manufactured meet increasing demand industry, commerce, and agriculture, among other areas. However, these types of buildings have not been focus any research field history, generating a lack knowledge regarding their types, distribution, related context enable statements on ability worthiness historical monument listings. This paper proposes methodology for automatic detection using...

Abstract Sea monitoring is essential for a better understanding of its dynamics and to measure the impact human activities. In this context, remote sensing plays an important role by providing satellite imagery every day, even in critical climate conditions, detection sea events with potential risk environment. The present work proposes comparative study Deep Learning architectures classification natural man-made using SAR imagery. evaluated comprises models based on convolutional networks,...

Per-point classification is a traditional method for remote sensing data classification, and radar in particular. Compared with optical data, the discriminative power of quite limited, most applications. A way trying to overcome these difficulties use Region-Based Classification (RBC), also referred as Geographical Object-Based Image Analysis (GEOBIA). RBC methods first aggregate pixels into homogeneous objects, or regions, using segmentation procedure. Moreover, known be an ill-conditioned...

Hyperspectral imaging is a technique in remote sensing that collect hundreds of images at differents wavelength values the same area Earth. For instance, Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS) sensor NASA capable to obtain 224 spectral channels range between 400 and 2500 nanometers. As result, each pixel image can be represented as signature. Image segmentation process dividing digital into groups pixels or objects. classification an important active dedicated identifying...

This work presents a spatio-temporal Conditional Random Field (CRF) based model for crop recognition from multi-temporal remote sensing image sequences. The association potential at each site is on the class posterior probabilities computed by Forest (RF) classifier given features corresponding site. A contrast-sensitive Potts used as label smoothing method in spatial domain, whereas interactions temporal domain are modeled expert knowledge about possible transitions between adjacent epochs....

Abstract. The presence of weeds in agricultural crops has been one the problems greatest interest recent years as they consume natural resources and negatively affect process. For this purpose, a model implemented to segment weed aerial images. proposed relies on DeepLabv3 architecture trained upon patches extracted from high-resolution imagery. dataset employed consisted 5 images that describes sugar beet field Germany. SegNet U-Net architectures were selected for comparison purposes. Our...

This work presents an approach for multispectral image classification that makes use of a Fuzzy Inference System (FIS). An IKONOS satellite sensor neighborhood in Rio de Janeiro, Brazil has been used. The ground truth used this comprises six classes: trees, scrub, buildings, roads, water and shadows. Then, inputs sets, rules outputs sets were defined. Four input variables, based on four indexes computed from the spectral bands, have considered: Normalized Difference Vegetation Index (NDVI),...

Motivated by the rapidly increase of remote sensing data over last years, this paper presents a tool designed for distributed imagesegmentation. InterSeg is able to handle efficiently very large high-resolution images using scalable segmentationmethods computer clusters. Through graphic user interface, hides processing complexity, allowing theuser easily perform image segmentation on cloud-computing environment. Experiments carried out with toolattest its potential scalability and capability...

In the case of tree species delineation with very high spatial resolution (VHR) images, is desirable that each segment corresponds to one individual crown (ITC). However, in order have a segmentation algorithm generates segments matching ITCs, its parameters ought be properly tuned. Aiming avoid time-consuming trial-and-error procedures associated this task, some initiatives for automatic search been developed, such as metaheuristic methods. The objective work was test tuning three...

Abstract. Applying remote sensing technology to map and monitor agriculture its impacts can greatly contribute for the proper development of this activity, promoting efficient food, fiber energy production. For that, not only images are needed, but also ground truth information, which is a key factor improvement methodologies using data. While variety current available, inclusive cost-free images, field reference data scarcer. agricultural applications, especially in tropical regions such as...

Abstract Container cranes are of key importance for maritime cargo transportation. The uninterrupted and all-day operation these container cranes, which directly affects the efficiency port, necessitates continuous inspection massive hoisting steel structures. Due to large size current manual inspections performed by expert climbers costly, risky, time-consuming. This motivates further investigations on automated non-destructive approaches remote fatigue-prone parts cranes. In this paper, we...

Abstract Accurate and up-to-date building road data are crucial for informed spatial planning. In developing regions in particular, major challenges arise due to the limited availability of these data, primarily as a result inherent inefficiency traditional field-based surveys manual generation methods. Importantly, this limitation has prompted exploration alternative solutions, including use remote sensing machine learning-generated (RSML) datasets. Within field RSML datasets, plethora...

Cracks are among the earliest indicators of deterioration in concrete structures. Early automatic detection these cracks can significantly extend lifespan critical infrastructures, such as bridges, buildings, and tunnels, while simultaneously reducing maintenance costs facilitating efficient structural health monitoring. This study investigates whether leveraging multi-temporal data for crack segmentation enhance quality. Therefore, we compare a Swin UNETR trained on with U-Net mono-temporal...