In comparison with conventional machine learning algorithms, deep learning can effectively express the deep features of remote sensing images. Considering the rich spectral and spatial information contained in hypersp...
详细信息
In comparison with conventional machine learning algorithms, deep learning can effectively express the deep features of remote sensing images. Considering the rich spectral and spatial information contained in hyperspectral images (HSIs), a combination method was proposed for HSI classification based on stacked autoencoder (SAE) and3ddeep residual network (3ddRN). Specifically, a SAE neuralnetwork was first built to reduce the dimensions of original HSIs. A 3d convolutional neural network (3dCNN) was then designed and the residual network module was introduced to build a 3ddRN. The dimension-reduced3d HSI cubes were input into the 3ddRN to extract identifiable joint spectral-spatial features. Finally, the deep features continuously identified by the 3ddRN were input to Softmax classification layer to realize the classification. In addition, Batch Normalization (BN) anddropout were usedduring the learning process to avoid overfitting on training data. The training and test sets of Indian Pines (IP), Pavia University (PU) and Salinas (SA) hyperspectral data sets were selected as the modeling and verification data sources. Six classical classification algorithms were adopted for comparing our proposed method, specifically including conventional machine learning algorithms of Radial Basis FunctionSupport Vector Machine (RBF-SVM), Kernel Simultaneous Orthogonal Matching Pursuit (KSOMP) and Local Binary Pattern-K-Nearest Neighbor (LBP-KNN), and mainstream deep learning algorithms of Variational Autoencoder (VAE), convolutionalneuralnetwork (CNN) and Spectral-Spatial Residual network (SSRN). The results showed that the overall accuracy (OA) reached 98.97%, 99.69% and 99.24%, respectively, only based on 10%, 5% and 1% of training samples for IP, PU and SA. Consequently, the proposed method shows a better classification performance, even in the case of limited samples.
Background: Sleep apnea is a respiratory disorder characterized by frequent breathing cessation during sleep. Sleep apnea severity is determined by the apnea-hypopnea index (AHI), which is the hourly rate of respirato...
详细信息
Background: Sleep apnea is a respiratory disorder characterized by frequent breathing cessation during sleep. Sleep apnea severity is determined by the apnea-hypopnea index (AHI), which is the hourly rate of respiratory events. In positional sleep apnea, the AHI is higher in the supine sleeping position than it is in other sleeping positions. Positional therapy is a behavioral strategy (eg, wearing an item to encourage sleeping toward the lateral position) to treat positional apnea. The gold standard of diagnosing sleep apnea and whether or not it is positional is polysomnography;however, this test is inconvenient, expensive, and has a long waiting list. Objective: The objective of this study was to develop and evaluate a noncontact method to estimate sleep apnea severity and to distinguish positional versus nonpositional sleep apnea. Methods: A noncontact deep-learning algorithm was developed to analyze infrared video of sleep for estimating AHI and to distinguish patients with positional vs nonpositional sleep apnea. Specifically, a 3d convolutional neural network (CNN) architecture was used to process movements extracted by optical flow to detect respiratory events. Positional sleep apnea patients were subsequently identified by combining the AHI information provided by the 3d-CNN model with the sleeping position (supine vs lateral) detected via a previously developed CNN model. Results: The algorithm was validated on data of 41 participants, including 26 men and 15 women with a mean age of 53 (Sd 13) years, BMI of 30 (Sd 7), AHI of 27 (Sd31) events/hour, and sleep duration of 5 (Sd 1) hours;20 participants had positional sleep apnea, 15 participants had nonpositional sleep apnea, and the positional status could not be discriminated for the remaining 6 participants. AHI values estimated by the 3d-CNN model correlated strongly and significantly with the gold standard (Spearman correlation coefficient 0.79, P<.001). Individuals with positional sleep apnea (based o
The reduction of visibility adversely affects land, marine, and air transportation. Thus, the ability to skillfully predict fog would provide utility. We predict fog visibility categories below 1600 m, 3200 m and 6400...
详细信息
The reduction of visibility adversely affects land, marine, and air transportation. Thus, the ability to skillfully predict fog would provide utility. We predict fog visibility categories below 1600 m, 3200 m and 6400 m by post -processing numerical weather prediction model output and satellite -based sea surface temperature (SST) using a 3d -convolutionalneuralnetwork (3d -CNN). The target is an airport located on a barrier island adjacent to a major US port;measured visibility from this airport serves as a proxy for fog that develops over the port. The features chosen to calibrate and test the model originate from the North American Mesoscale Forecast System, with values of each feature organized on a 32 x 32 horizontal grid;the SSTs were obtained from the NASA Multiscale Ultra Resolution dataset. The input to the model is organized as a high dimensional cube containing 288 to 384 layers of 2d horizontal fields of meteorological variables (predictor maps). In this 3d -CNN (hereafter, FogNet), two parallel branches of feature extraction have been designed, one for spatially auto -correlated features (spatial -wise dense block and attention module), and the other for correlation between input variables (variable -wise dense block and attention mechanism.) To extract features representing processes occurring at different scales, a 3d multiscale dilated convolution is used. data from 2009 to 2017 (2018 to 2020) are used to calibrate (test) the model. FogNet performance results for 6, 12- and 24 - h lead times are compared to results from the High -Resolution Ensemble Forecast (HREF) system. FogNet outperformed HREF using 8 standard evaluation metrics.
In recent years, the proportion of deaths from cancer tends to increase in Japan, especially the number of deaths from lung cancer is increasing. CT device is effective for early detection of lung cancer. However, the...
详细信息
ISBN:
(纸本)9781538670804;9788993215168
In recent years, the proportion of deaths from cancer tends to increase in Japan, especially the number of deaths from lung cancer is increasing. CT device is effective for early detection of lung cancer. However, there is concern that an increase in burden on doctors will be caused by high performance of CT improving. Therefore, by presenting the "second opinion" by the CAd system, it reduces the burden on the doctor. In this paper, we develop a CAd system for automatic detection of lesion candidate regions such as lung nodules or ground glass opacity (GGO) from 3d CT images. Our proposed method consists of three steps. In the first step, lesion candidate regions are extracted using temporal subtraction technique. In the second step, the image is reconstructed by sparse coding for the extracted region. In the final step, 3d convolutional neural network (3d-CNN) identification using reconstructed images is performed. We applied our method to 51 cases and True Positive rate (TP) of 79.81% and False Positive rate (FP) of 37.65% are obtained.
The problem of class imbalance exists in detecting the pulmonary nodules from Computed Tomography(CT) by means of convolutionalneuralnetwork. A Three-dimensional detector Based on Focal Loss(FLTdd) is designed in th...
详细信息
The problem of class imbalance exists in detecting the pulmonary nodules from Computed Tomography(CT) by means of convolutionalneuralnetwork. A Three-dimensional detector Based on Focal Loss(FLTdd) is designed in this paper to ensure that the pulmonary nodules in CT could be identified more exactly. Its framework focuses more on samples that are difficult to be classified. Besides, three dimensional detector contains richer spatial information and gets more distinguishing features. The experiment results obtained from LIdC-IdRI data set show that the average sensitivity score of FLTdd achieves89.62%. It has a 1.47% improvement compared with the published CASEd method.
This study presents a vision-based human action recognition system using a deep learning technique. The system can recognize human actions successfully when the camera of a robot is moving toward the target person fro...
详细信息
ISBN:
(纸本)9781450372213
This study presents a vision-based human action recognition system using a deep learning technique. The system can recognize human actions successfully when the camera of a robot is moving toward the target person from various directions. Therefore, the proposed method is useful for the vision system of indoor mobile robots. The system uses three types of information to recognize human actions, namely, information from color videos, optical flow videos, anddepth videos. First, Kinect 2.0 captures color videos anddepth videos simultaneously using its RGB camera anddepth sensor. Second, the histogram of oriented gradient features is extracted from the color videos, and a support vector machine is used to detect the human region. Based on the detected human region, the frames of the color video are cropped and the corresponding frames of the optical flow video are obtained using the Farnebäck method (https://***=.org/3.4/d4/dee/ tutorial_optical_***). The number of frames of these videos is then unified using a frame sampling technique. Subsequently, these three types of videos are input into three modified3d convolutional neural networks (3d CNNs) separately. The modified3d CNNs can extract the spatiotemporal features of human actions and recognize them. Finally, these recognition results are integrated to output the final recognition result of human actions. The proposed system can recognize 13 types of human actions, namely, drink (sit), drink (stand), eat (sit), eat (stand), read, sit down, stand up, use a computer, walk (horizontal), walk (straight), play with a phone/tablet, walk away from each other, and walk toward each other. The average human action recognition rate of 369 test human action videos was 96.4%, indicating that the proposed system is robust and efficient.
The novel COVId-19 is a global pandemic disease overgrowing worldwide. Computer-aided screening tools with greater sensitivity are imperative for disease diagnosis and prognosis as early as possible. It also can be a ...
详细信息
暂无评论