Accurate segmentation of uterus, uterine fibroids, and spine from MR images is crucial for high intensity focused ultrasound (HIFU) therapy but remains still difficult to achieve because of 1) the large shape and size...
详细信息
Accurate segmentation of uterus, uterine fibroids, and spine from MR images is crucial for high intensity focused ultrasound (HIFU) therapy but remains still difficult to achieve because of 1) the large shape and size variations among individuals, 2) the low contrast between adjacent organs and tissues, and 3) the unknown number of uterine fibroids. To tackle this problem, in this paper, we propose a large kernel encoder-decoder Network based on a 2D segmentation model. The use of this large kernel can capturemulti-scale contexts by enlarging the valid receptive field. In addition, a deep multiple atrous convolution block is also employed to enlarge the receptive field and extract denser feature maps. Our approach is compared to both conventional and other deep learning methods and the experimental results conducted on a large dataset show its effectiveness.
Semantic segmentation of 3D point clouds is a crucial task in scene understanding and is also fundamental to indoor scene applications such as indoor navigation, mobile robotics, augmented reality. Recently, deep lear...
详细信息
Semantic segmentation of 3D point clouds is a crucial task in scene understanding and is also fundamental to indoor scene applications such as indoor navigation, mobile robotics, augmented reality. Recently, deep learning frameworks have been successfully adopted to point clouds but are limited by the size of data. While most existing works focus on individual sampling points, we use surface patches as a more efficient representation and propose a novel indoor scene segmentation framework called patch graph convolution network (PGCNet). This framework treats patches as input graph nodes and subsequently aggregates neighboring node features by dynamic graph U-Net (DGU) module, which consists of dynamic edge convolution operation inside U-shaped encoder-decoder architecture. The DGU module dynamically update graph structures at each level to encode hierarchical edge features. Incorporating PGCNet, we can segment the input scene into two types, i.e., room layout and indoor objects, which is afterward utilized to carry out final rich semantic labeling of various indoor scenes. With considerable speedup training, the proposed framework achieves effective performance equivalent to state-of-the-art for segmenting standard indoor scene dataset.
With the rapid development of the convolutional neural network, both instance segmentation and semantic segmentation have achieved remarkable performances. Recently, many efforts have been made to use a unified Encode...
详细信息
With the rapid development of the convolutional neural network, both instance segmentation and semantic segmentation have achieved remarkable performances. Recently, many efforts have been made to use a unified encoder-decoder architecture to solve these two segmentation tasks simultaneously. The encoder extracts high-level features from the input images for both tasks. However, existing decoders cannot meet the performance requirements of these two tasks: the semantic segmentation decoder is not flexible enough for instance segmentation, and the instance segmentation decoder lacks the precision of semantic segmentation. Therefore, we introduce a novel Pixel Voting decoder to satisfy both precision and flexibility. The proposed decoder regresses the interlayer pixel relationships between the input and output feature maps across the convolutional layers. Then, the pixel relationships are regarded as the pixel votes for dynamically decoding the higher level information from the encoder. Finally, we propose the dynamic deconvolution to make full use of the votes for each pixel during the decoding process. Meanwhile, the matrix computation for the dynamic deconvolution is designed to boost the calculation. Experiments show that the proposed method can achieve better performance than the well-known methods on both instance segmentation on the COCO dataset and semantic segmentation on the Cityscapes dataset. The matrix implementation of the dynamic deconvolution also shows its high efficiency and feasibility.
Neuromorphic computing has recently gained significant attention as a promising approach for developing energy-efficient, massively parallel computing systems inspired by the spiking behavior of the human brain and na...
详细信息
Neuromorphic computing has recently gained significant attention as a promising approach for developing energy-efficient, massively parallel computing systems inspired by the spiking behavior of the human brain and natively mapping spiking neural networks (SNNs). Effective training algorithms for SNNs are imperative for increased adoption of neuromorphic platforms;however, SNN training continues to lag behind advances in other classes of ANN. In this paper, we reduce this gap by proposing an innovative encoder-decoder technique that leverages sparse coding and the locally competitive algorithm (LCA) to provide an algorithm specifically designed for neuromorphic platforms. Using our proposed Dataset-Scalable Exemplar LCA-decoder we reduce the computational demands and memory requirements associated with training SNNs using error backpropagation methods on increasingly larger training sets. We offer a solution that can be scalably applied to datasets of any size. Our results show the highest reported top-1 test accuracy using SNNs on the ImageNet and CIFAR100 datasets, surpassing previous benchmarks. Specifically, we achieved a record top-1 accuracy of 80.75% on ImageNet (ILSVRC2012 validation set) and 79.32% on CIFAR100 using SNNs.
Technology that translates photoplethysmogram (PPG) into the QRS complex of electrocardiogram (ECG) would be transformative for people who require continuously monitoring. However, directly decoding the QRS complex of...
详细信息
Technology that translates photoplethysmogram (PPG) into the QRS complex of electrocardiogram (ECG) would be transformative for people who require continuously monitoring. However, directly decoding the QRS complex of ECG from PPG is challenging because PPG signals usually have different offsets due to 1) different devices, and 2) personal differences, which makes the alignment difficult. In this paper, we make the first attempt to reconstruct the QRS complex of ECG only from the recording of PPG by an end-to-end deep learning-based approach. Specifically, we propose a novel encoder-decoder architecture containing three components: 1) a sequence transformer network which automatically calibrates the offset, 2) an attention network, which dynamically identifies regions of interest, and 3) a new QRS complex-enhanced loss for better reconstruction. The experiment results on a real dataset demonstrate the effectiveness of the proposed method: 3.67% R peak failure rate of the reconstructed ECG and high correlation of pulse transit time between the reconstructed QRS complex and the groundtruth QRS complex (rho = 0.844), which creates a new opportunity for low-cost clinical studies via the waveform-level reconstruction of the QRS complex of ECG from PPG.
In this paper, the tracking performance limitation of networked control systems (NCSs) is studied. The NCSs are considered as continuous-time linear multi-input multioutput (MIMO) systems with random reference noises....
详细信息
In this paper, the tracking performance limitation of networked control systems (NCSs) is studied. The NCSs are considered as continuous-time linear multi-input multioutput (MIMO) systems with random reference noises. The controlled plants include unstable poles and nonminimum phase (NMP) zeros. The output feedback path is affected by multiple communication constraints. We focus on some basic communication constraints, including additive white noise (AWN), quantization noise, bandwidth, as well as encoder-decoder. The system performance is evaluated with the tracking error energy, and used a two-degree-of-freedom (2DOF) controller. The explicit representation of the tracking performance is given in this paper. The results indicate the tracking performance limitations rely to internal characteristics of the plant (unstable poles and NMP zeros), reference noises [the reference noise power distribution (RNPD) and its directions], and the characteristics of communication constraints. The characteristics of communication constraints include communication noise power distribution (CNPD);quantization noise power distribution (QNPD), and their distribution directions;transform bandwidth allocation (TBA);transform encoder-decoder allocation (TEA), and their allocation directions;and NMP zeros and MP part of bandwidth. Moreover, the tracking performance limitations are also affected by the angles between the each transform NMP zero direction and RNPD direction, and these angles between each transform unstable poles direction and the direction of communication constraint distribution/allocation. In addition, for MIMO NCSs, bandwidth (there are not identical two channels) can always affect the direction of unstable poles, and the channel allocation of bandwidth and encode-decode may be used for a feasible method for the performance allocation of each channel. Finally, an instance is given for verifying the effectiveness of the theoretical outcomes.
Accurate multi-energy load forecasting (MELF) is the key to realize the balance between supply and demand in regional integrated energy systems (RIES). To this end, a hybrid MELF method for RIES considering temporal d...
详细信息
Accurate multi-energy load forecasting (MELF) is the key to realize the balance between supply and demand in regional integrated energy systems (RIES). To this end, a hybrid MELF method for RIES considering temporal dynamic and coupling characteristics (MELF_TDCC) is proposed. The novelty of MELF_TDCC lies in the following three aspects: 1) considering the high-dimensional temporal dynamic characteristic, an encoder-decoder model based on long-short term memory network (LSTMED) is proposed, which can extract the high dimensional potential feature, and reflect the temporal dynamic characteristics of historical load sequence effectively;2) considering the cross-coupling characteristic, a coupling feature matrix of multi-energy load is constructed, which reflects the cross-influence of electricity, cooling and heating loads;3) with the feature fusion layer of the hybrid model being built by gradient boosting decision tree (GBDT), the extended feature matrix for each class of load is constructed considering the intra-class inherent characteristics and inter-class coupling characteristic of loads, and the GBDT model is trained on the extended feature matrix, which provides multi-dimensional perspective for researching load essential characteristics. MELF_TDCC is verified on the ultra-short-term and short-term MELF scenarios based on an actual dataset. The simulation result shows that the proposed MELF_TDCC outperforms the current advanced methods. (C) 2020 Elsevier Ltd. All rights reserved.
作者:
Zhu, DiCheng, XimengZhang, FanYao, XinGao, YongLiu, YuPeking Univ
Sch Earth & Space Sci Inst Remote Sensing & Geog Informat Syst Beijing Peoples R China Peking Univ
Beijing Key Lab Spatial Informat Integrat & Its A Beijing Peoples R China UCL
SpaceTimeLab Dept Civil Environm & Geomat Engn London England MIT
Senseable City Lab 77 Massachusetts Ave Cambridge MA 02139 USA
Spatial interpolation is a traditional geostatistical operation that aims at predicting the attribute values of unobserved locations given a sample of data defined on point supports. However, the continuity and hetero...
详细信息
Spatial interpolation is a traditional geostatistical operation that aims at predicting the attribute values of unobserved locations given a sample of data defined on point supports. However, the continuity and heterogeneity underlying spatial data are too complex to be approximated by classic statistical models. Deep learning models, especially the idea of conditional generative adversarial networks (CGANs), provide us with a perspective for formalizing spatial interpolation as a conditional generative task. In this article, we design a novel deep learning architecture named conditional encoder-decoder generative adversarial neural networks (CEDGANs) for spatial interpolation, therein combining the encoder-decoder structure with adversarial learning to capture deep representations of sampled spatial data and their interactions with local structural patterns. A case study on elevations in China demonstrates the ability of our model to achieve outstanding interpolation results compared to benchmark methods. Further experiments uncover the learned spatial knowledge in the model's hidden layers and test the potential to generalize our adversarial interpolation idea across domains. This work is an endeavor to investigate deep spatial knowledge using artificial intelligence. The proposed model can benefit practical scenarios and enlighten future research in various geographical applications related to spatial prediction.
The single imbalance pricing is an emerging mechanism in European electricity markets where all positive and negative imbalances are settled at a unique price. This real-time scheme thereby stimulates market participa...
详细信息
The single imbalance pricing is an emerging mechanism in European electricity markets where all positive and negative imbalances are settled at a unique price. This real-time scheme thereby stimulates market participants to deviate from their schedule to restore the power system balance. However, exploiting this market opportunity is very risky due to the extreme volatility of the real-time power system conditions. In order to address this issue, we implement a new tailored deep-learning model, named encoder-decoder, to generate improved probabilistic forecasts of the imbalance signal, by efficiently capturing its complex spatio-temporal dynamics. The predicted distributions are then used to quantify and optimize the risk associated with the real-time participation of market players, acting as price-makers, in the imbalance settlement. This leads to an integrated forecast-driven strategy, modeled as a robust bi-level optimization. Results show that our probabilistic forecaster achieves better performance than other state of the art tools, and that the subsequent risk-aware robust dispatch tool allows finding a tradeoff between conservative and risk-seeking policies, leading to improved economic benefits. Moreover, we show that the model is computationally efficient and can thus be incorporated in the very-short-term dispatch of market players with flexible resources.
暂无评论