Synthetic aperture radar (SAR) image classification is a hot topic in the interpretation of SAR images. However, absence effective feature representation and presence speckle noise images make difficult to handle. In order overcome these problems, deep convolutional autoencoder (DCAE) proposed extract features conduct automatically. The network composed eight layers: layer texture features, scale transformation aggregate neighbor information, four layers based on sparse autoencoders optimize...

Deep learning, which represents data by a hierarchical network, has proven to be efficient in computer vision. To investigate the effect of deep features hyperspectral image (HSI) classification, this paper focuses on how extract and utilize HSI classification framework. First, order spectral-spatial information, an improved spatial updated auto-encoder (SDAE), is proposed. SDAE, (DAE), considers sample similarity adding regularization term energy function, updates integrating contextual...

Few-shot learning in image classification is developed to learn a model that aims identify unseen classes with only few training samples for each class. Fewer and new tasks of make many traditional models no longer applicable. In this paper, novel few-shot method named multi-scale metric (MSML) proposed extract features the relations between learning. method, feature pyramid structure introduced embedding, which combine high-level strong semantic low-level but abundant visual features. Then...

The existing deep networks have shown excellent performance in remote sensing scene classification (RSSC), which generally requires a large amount of class-balanced training samples. However, will result underfitting with imbalanced samples since they can easily bias toward the majority classes. To address these problems, multigranularity decoupling network (MGDNet) is proposed for image classification. begin with, we design complementary feature representation (MGCFR) method to extract...

Abstract Because the reliability of feature for every pixel determines accuracy classification, it is important to design a specialized mining algorithm hyperspectral image classification. We propose learning algorithm, contextual deep learning, which extremely effective On one hand, learning-based extraction can characterize information better than pre-defined algorithm. other spatial Contextual explicitly learns spectral and features via architecture promotes extractor using supervised...

The classification of a synthetic aperture radar (SAR) image is significant yet challenging task, due to the presence speckle noises and absence effective feature representation. Inspired by deep learning technology, novel supervised contractive neural network (DSCNN) for SAR proposed overcome these problems. In order extract spatial features, multiscale patch-based extraction model that consists gray level-gradient co-occurrence matrix, Gabor, histogram oriented gradient descriptors...

Change detection is an important task to identify land-cover changes between the acquisitions at different times. For synthetic aperture radar (SAR) images, inherent speckle noise of images can lead false changed points, which affects change performance. Besides, supervised classifier in framework requires numerous training samples, are generally obtained by manual labeling. In this paper, a novel unsupervised method named saliency-guided deep neural networks (SGDNNs) proposed for SAR image...

The problem of different characters heterogeneous synthetic aperture radar (SAR) images leads to poor performances for transfer learning SAR image classification. To address this issue, a semisupervised model named as deep joint distribution adaptation networks (DJDANs) is proposed from source but similar target image, which aims match the probability distributions between domain and domain. In DJDAN, marginal network developed map features across domains into an augmented common feature...

Deep learning has achieved excellent performance in remote-sensing image scene classification, since a large number of datasets with annotations can be applied for training. However, actual applications, there is just few annotated samples and unannotated images, which leads to overfitting the deep model affects classification. In order address these problems, semi-supervised representation consistency Siamese network (SS-RCSN) proposed First, considering intraclass diversity interclass...

Synthetic aperture radar (SAR) image classification is a fundamental process for SAR understanding and interpretation. With the advancement of imaging techniques, it permits to produce higher resolution data extend amount. Therefore, intelligent algorithms high-resolution are demanded. Inspired by deep learning technology, an end-to-end model from original final map developed automatically extract features conduct classification, which named recurrent encoding neural networks (DRENNs). In...

Abstract Background Sepsis-associated encephalopathy (SAE) is related to increased short-term mortality in patients with sepsis. We aim establish a user-friendly nomogram for individual prediction of 30-day risk SAE. Methods Data were retrospectively retrieved from the Medical Information Mart Intensive Care (MIMIC III) open source clinical database. SAE was defined by Glasgow Coma Score (GCS) < 15 or delirium at presence Prediction model constructed training set logistic regression...

Semi-supervised few-shot learning is developed to train a classifier that can adapt new tasks with limited labeled data and fixed quantity of unlabeled data. Most semi-supervised methods select pseudo-labeled set by task-specific confidence estimation. This work presents task-unified estimation approach for learning, named pseudo-loss metric (PLCM). It measures the credibility loss distribution pseudo-labels, which synthetical considered multi-tasks. Specifically, different are mapped...

Deep learning algorithms that can effectively extract features from different modalities have achieved significant performance in multimodal remote sensing (RS) data classification. However, we actually found the feature representation of one modality is likely to affect other through parameter back-propagation. Even if models are superior their uni-modal counterparts, they be underutilized. To solve above issue, a dual-branch dynamic modulation network proposed for hyperspectral (HS) and...

Polarimetric synthetic aperture radar (PolSAR) has attracted more attentions because of its excellent observation ability, and PolSAR image classification become one the significant tasks in remote sensing interpretation. Various types sizes land cover objects lead to misclassification, especially boundaries different categories. To solve these issues, a multiscale superpixel-guided weighted graph convolutional network (MSGWGCN) is proposed for classifying images. In MSGWGCN, superpixel...

Digital twin models are computerized clones of physical assets or systems and have attracted much attention from academia industries. applications focus on smart manufacturing systems. Meanwhile, products driven increasingly by the needs customers. Industrial production modes evolved mass to personalized production. Understanding customers meeting their become important issues in manufacturing. Social networks provide platforms for online engage different behaviors. In addition, personality...

Few-shot classification of remote sensing images has attracted attention due to its important applications in various fields. The major challenge few-shot image scene is that limited labeled samples can be utilized for training. This may lead the deviation prototype feature expression, and thus performance will impacted. To solve these issues, a calibration with feature-generating model proposed classification. In framework, encoder self-attention developed reduce influence irrelevant...

Remote sensing image semantic segmentation (RSISS) remains challenging due to the scarcity of labeled data. Semi-supervised learning can leverage pseudo-labels enhance model's ability learn from unlabeled However, accurately generating for RSISS a significant challenge that severely affects performance, especially edges different classes. In order overcome these issues, we propose semi-supervised framework remote images based on edge-aware class activation enhancement (ECAE). Firstly,...