Shadow removal is an essential task for scene understanding. Many studies consider only matching the image contents, which often causes two types of ghosts: color in-consistencies in shadow regions or artifacts on boundaries (as shown Figure. 1). In this paper, we tackle these issues ways. First, to carefully learn border artifacts-free image, propose a novel network structure named dual hierarchically aggregation (DHAN). It contains series growth dilated convolutions as backbone without any...

Change detection with heterogeneous remote sensing images (HRSIs) is attractive for observing the Earth's surface when homogeneous are unavailable. However, HRSIs cannot be compared directly because imaging mechanisms bitemporal different, and detecting change challenging. In this letter, a simple yet effective deep learning approach based on classical UNet proposed. First, pair of image patches concatenated together to learn shared abstract feature in both patch domains. Then, multiscale...

Land cover change detection (LCCD) based on bitemporal remote sensing images has become a popular topic in the field of sensing. Despite numerous methods promoted recent decades, an improvement usability and performance these remained necessary. In this paper, novel LCCD approach integration k-means clustering adaptive majority voting (kmeans_AMV) techniques have been developed. The proposed k-means_AMV method consists three major techniques. First, to utilize contextual information manner,...

This paper develops a novel hyperspectral image (HSI) classification framework by exploiting the spectral-spatial features of multiscale superpixels via recurrent neural networks with stacked autoencoders. The can be used to segment an HSI into shape-adaptive regions, and capture object information more accurately. Therefore, superpixel-based methods have been studied many researchers. In this paper, we propose method. contrast current research, proposed method not only captures each scale...

The impact of urban noise on residents' physical health and mental condition has gradually become a hot topic public discussion. A single monitoring cannot fully reflect the public's actual perception feeling noise. In contrast, social data can provide indirect reference for This study first collected synthesized complaint from government platform. Furthermore, spatial temporal distribution characteristics complaints in blocks were analyzed to explore relationship between behaviours Point...

Land cover change detection (LCCD) with very high-resolution remote-sensing images (VHR_RSIs) is important in observing surface on Earth. However, pseudo changes usually reduces the accuracy of map. In this paper, novel piecewise distance based adaptive region key-points extraction called sparse key-point (SKPD) developed to measure magnitude between bitemporal VHR_RSIs for LCCD. The proposed approach consists three steps. First, an generation algorithm promoted exploring spatial-contextual...

Infection, inflammation, tumors, and other illnesses have for a long time posed major danger to human health developed into global public issue. However, the existing treatment methods are sometimes not effective, leading high mortality. Therefore, it is urgent seek new alternative therapies. Gallium related derivatives received lot of interest recently because their exceptional biological safety distinctive pharmacological properties. More crucially, gallium its compounds potential...

Vision Transformer, which performs well in various vision tasks, encounters a bottleneck skeleton-based action recognition and falls short of advanced GCN-based methods. The root cause is that the current skeleton transformer depends on self-attention mechanism complete channel global joint, ignoring highly discriminative differential correlation within channel, so it challenging to learn expression multivariate topology dynamically. To tackle this, we present Skeleton MixFormer, an...

Constructing a resilient ecological network (EN) and identifying critical strategic nodes corridors within the EN, stands as pivotal approach toward achieving harmonious equilibrium between regional development conservation. It is imperative to expand perspective encompass holistic resilience of thus safeguarding ecologically that play roles throughout entire network. Therefore, this study takes Nanjing City case establishes comprehensive EN assessment framework based on theory complex...

Emotion plays a nuclear part in human attention, decision-making, and communication. Electroencephalogram (EEG)-based emotion recognition has developed lot due to the application of Brain-Computer Interface (BCI) its effectiveness compared body expressions other physiological signals. Despite significant progress affective computing, is still an unexplored problem. This paper introduced Logistic Regression (LR) with Gaussian kernel Laplacian prior for EEG-based recognition. The enhances EEG...

Natural soundscape is considered a dominant type of hearing in forested areas and contributes to health recovery effects from exposure the biophilic outdoor environment. This study focuses on different forest structures, aims explore relationship between perceived acoustical parameters, observe physiological indicators, model restorative role soundscape. Questionnaires measuring equipment were used gather psychophysical information at 20 observation sites urban areas. Back-propagation neural...

Change detection with remote sensing images (RSIs) plays an important role in the community of applications. However, when change is conducted hyperspectral (HRSIs), how to measure magnitude between bitemporal HRSIs becomes challenging due high dimension HRSIs. In this article, a novel Distribution Distance based on Inconsistent Adaptive Region (D <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> IAR) approach proposed for improving...

Self-supervised contrastive learning is a powerful pre-training framework for the invariant features from different views of remote sensing images, therefore, performance heavily depends on generation views. Current view primarily accomplished through transformations, and types parameters transformations are require hand-crafted. Hence, diversity discriminability generated cannot be guaranteed. To address this, we propose multi-type optimization method to optimize these transformations. We...

Urban forests by the riverside are important habitats for various animals and contribute soundscapes citizens. Unfortunately, urban exposed to influence of traffic noises from freeways. This study aims explore spatial temporal variation soundscape, conduct soundscape optimization multiple parameters, find a balance its interval elements through optimizing map. Questionnaires measuring equipment were used gather information in an forested area Fuzhou, China. Diurnal variations mapping analyze...

Physical activity brings multiple health benefits to seniors. Neighborhood parks provide seniors with accessible spaces and opportunities engage in physical activity. This study investigated the associations between neighborhood park design characteristics seniors' total walking step energy expenditure during visit. Seniors' was measured by pedometer, calculated based on self-reported activities park. The conducted 15 an area less than 10 ha, included 234 senior participants. ANOVA analyses...

This article presents a novel dual-path full convolutional network (DP-FCN) model for constructing landslide inventory map (LIM) with bitemporal very high-resolution (VHR) remote sensing images. Unlike traditional methods drawing LIM, the proposed DP-FCN directly draws LIMs from aerial images VHR through trained deep neural without generating change magnitude map. Thus, approach can effectively reduce effects of pseudo changes caused by phenological differences rather than events. The...