Road pavement cracks automated detection is one of the key factors to evaluate road distress quality, and it a difficult issue for construction intelligent maintenance systems. However, has been challenging task, including strong nonuniformity, complex topology, noise-like problems in crack images, so on. To address these challenges, we propose CrackSeg—an end-to-end trainable deep convolutional neural network detection, which effective achieving pixel-level, via high-level features. In this...

Pavement crack detection and characterization is a fundamental part of road intelligent maintenance systems. Due to the high non-uniformity cracks, topological complexity, similar noise from texture, challenge arises in this domain with automated classification complex environment. In work, an overarching framework for universal robust automatic method that simultaneously characterizes type its severity level was developed. For detection, we propose novel efficient network captures context...

Treating each intersection as basic agent, multi-agent reinforcement learning (MARL) methods have emerged the predominant approach for distributed adaptive traffic signal control (ATSC) in multi-intersection scenarios, such arterial coordination. MARL-based ATSC currently faces two challenges: disturbances from policies of other intersections may impair and stability agents; heterogeneous features across complicate coordination efforts. To address these challenges, this study proposes a...

The distribution network reconfiguration (DNR) aims at minimizing the power losses and improving voltage profile. Traditional model-based methods exactly need parameters to derive optimal configuration of network. This paper proposes a DNR method based on model-free reinforcement learning (RL) approach. proposed adopts NoisyNet deep Q-learning (DQN), by which exploration can be automatically realized without tuning parameters, in order accelerate training process improve optimization...

Abstract Rapid building damage assessment following an earthquake is important for humanitarian relief and disaster emergency responses. In February 2023, two magnitude-7.8 earthquakes struck Turkey in quick succession, impacting over 30 major cities across nearly 300 km. A comprehensive understanding of the distribution essential efficiently deploying rescue forces during critical periods. This article presents training a two-stage convolutional neural network called BDANet that integrated...

Flood hazards resulting from short-term severe precipitation have caused serious social and economic losses posed extraordinary threats to the safety of lives property. Vulnerability, which reflects degree adverse impact flooding on a city, sensitivity environment, extent rescues are possible during flooding, is one significant factors disaster risk assessment. Because this, this paper proposes an Environmental Vulnerability Analysis Model (EVAM), based comprehensively evaluating...

Fast‐match is a fast and effective algorithm for template matching. However, when matching colour images, the images are converted into greyscale images. The information lost in this process, resulting errors areas with distinctive colours but similar values An improved fast‐match that utilises all three RGB channels to construct sum‐of‐absolute‐differences (CSAD) proposed, thus improving distance used fast‐match. In algorithm, each pixel image categorised by clustering them using...

Artificial Neural Networks (ANNs), as a nonlinear and adaptive information processing systems, play an important role in machine learning, artificial intelligence, data mining. But the performance of ANNs is sensitive to number neurons, chieving better network simplifying topology are two competing objectives. While Genetic Algorithms (GAs) kind random search algorithm which simulates nature selection evolution, has advantages good global abilities learning approximate optimal solution...

Accurate and timely risk assessment of short-term rainstorm-type flood disasters is very important for ecological environment protection sustainable socio-economic development. Given the complexity variability different geographical environments climate conditions, a single machine learning model may lead to overfitting issues in disaster assessment, limiting generalization ability such models. In order overcome this challenge, study proposed rainstorm framework under integrated model, which...

The <i>ZiYuan3</i> (<i>ZY3</i>) is a civilian stereo surveying and mapping satellite from China operating under the framework of Earth resources series, its objective to fulfill 1:50,000 update largerscale fundamental geographic information products. This article introduces <i>ZY3-03</i> satellite's mission payload specifications as well data utilization distribution policy. In future, will realize demands China's national economic social development play an important role in fields...

Climate change has led to an increased frequency of extreme precipitation events, resulting in damage from rainstorms and floods. Rapid efficient flood forecasting is crucial. However, traditional hydrological simulation methods that rely on site distribution are limited by the availability data cannot provide fast accurate monitoring information. Therefore, this study took event Huoqiu County 2020 as example proposes a three-dimensional method based active passive satellites, which provides...

As an essential technology for intelligent transportation management and traffic risk prevention control, vehicle detection plays a significant role in the comprehensive evaluation of system. However, limited by small size vehicles satellite remote sensing images lack sufficient texture features, its performance is far from satisfactory. In view unclear edge structure objects super-resolution (SR) reconstruction process, deep convolutional neural networks are no longer effective extracting...

The random noises caused by different devices in the process of X-ray imaging make images degraded, which results incomplete or even incorrect medical diagnoses. image pre-processing is primary procedure digital device. Due to features image, temporal recursive filter can be used, whose filtering coefficient decreases exponentially with difference two adjacent frames. However, considering hardware implementation complexity real-time dynamic processing, an improved self-adaptive algorithm...

How to select the suitable parameters and kernel model is a very important problem for Twin Support Vector Machines (TSVMs). In order solve this problem, one solving algorithm called Invasive Weed Optimization Algorithm Optimizating Parameters of Mixed Kernel (IWO-MKTSVMs) proposed in paper. Firstly, introducing mixed kernel, twin support vector machines based on constructed. This strategy good way selection. selection which contain TSVMs parameters, (IWO) introduced. IWO an optimization who...

Abstract. Super-resolution reconstruction of sequence remote sensing image is a technology which handles multiple low-resolution satellite images with complementary information and obtains one or more high resolution images. The cores the are precision matching between detail extraction fusion. In this paper puts forward new super model frame can adaptive multi-scale enhance details reconstructed image. First, were decomposed into layer containing smooth large scale edge by bilateral filter....

Recent convolutional neural networks have made significant advancements in the detection of road cracks. However, lack accurate crack training data reduces generalisation ability deep model. In this Letter, a semi‐automatic pavement labelling algorithm is proposed to solve problem insufficient data. First, modified C–V model used obtain preliminary segmentation results. Second, direction initial area calculated by ellipse fitting method, and results are as samples for labelling. Finally,...

Scene representation in the bird's-eye-view (BEV) coordinate frame provides a succinct and effective way to understand surrounding environments for autonomous vehicles robotics. In this work, we present an end-to-end architecture generate BEV from cameras. To representation, propose transformer-based encoder-decoder structure translate image features different cameras into frame, which takes advantage of context information individual relationship between images views. We perform multiple...

Vivid main structure and rich texture detail are important factors with which to determine the quality of high-resolution images after super-resolution (SR) reconstruction. Owing loss high-frequency information in process SR reconstruction limitation accurate estimation unknown inversion process, a gap still exists between image real image. The can better preserve edge image, boosting compensate for missing process. Therefore, novel single remote-sensing method based on multilevel (MMSDB-SR)...

To validate the feasibility of ZY-3-02 laser altimetry data in enhancing vertical accuracy domestic satellite stereo-imagery, this paper presents a composite geolocating method which adopts block adjustment model based on RFM (rational function model) with HCPs(horizontal control points) and ECPs(elevation separately. The two scenes optical stereo-imageries are deployed for validation. results show that absolute stereo-imagery instead field-surveyed GCPs (ground can be improved to about 3 meter.

Due to the influence of global climate anomalies, abnormal weather conditions such as heavy rainfall have become more frequent in recent years, posing a significant threat operation transportation systems. An effective assessment resilience system before and during rain can alert department take necessary emergency actions. However, existing methods for assessing networks mostly suffer from following problems: 1) Simulation modeling impacts lack realism; 2) After-the-fact evaluations cannot...