Different compilers and optimization levels can be used to compile the source code. Revealed in reverse from the produced binaries, these compiler details facilitate essential binary analysis tasks, such as malware fo...
详细信息
Convectional weather is one of the weather which often occurs during the warm season, the scope of this kind of weather is normally small, short duration, etc, so it is difficult to predicted, the effects of convectio...
详细信息
ISBN:
(纸本)9781450387828
Convectional weather is one of the weather which often occurs during the warm season, the scope of this kind of weather is normally small, short duration, etc, so it is difficult to predicted, the effects of convectional weather are very large, from the national economy to military activities will be threatened by it, so how to achieve more accurate of convection weather forecast will be of great significance. Doppler radar is one of the main tools for monitoring and early warning severe convective weather. It can provide important strong convective information for prediction. through its observed real-time echo intensity (reflectivity factor Z), radial velocity (V) and velocity spectrum width (W). Echo intensity is an important basis for judging whether severe, convective weather occurs, and radial velocity can identify wind damage and is also an important basis for judging the occurrence of severe convective weather. From the radial velocity, we can see the convergence, divergence and rotation characteristics of airflow, which are closely related to the occurrence and development of severe convective weather. Therefore, Doppler radar data plays a very important role in the detection and prediction of severe weather. At present, there are still problems about how to effectively extract these information from Doppler radar data. This paper proposes a method of near prediction based on deep learning. The neural network is used to automatically learn data features for prediction. Convolution neural network is combined with long-short-time memory network. Firstly, a three-dimensional convolution is proposed to extract the spatial features of three-dimensional original data, and then the long-short-time memory network is used to extract the features of data in time dimension, thus improving the prediction accuracy. Deep learning method avoids the process of manually extracting data features, and uses historical 3D Doppler radar imagedata to predict convective weather f
As a major form of data, the collection and analysis of time series have been widely used in many fields. In practice, wireless sensor network is a popular mechanism for data collection and it is suitable for time ser...
详细信息
ISBN:
(数字)9781728189543
ISBN:
(纸本)9781728189550
As a major form of data, the collection and analysis of time series have been widely used in many fields. In practice, wireless sensor network is a popular mechanism for data collection and it is suitable for time series. However, due to the diversity deployment area for a wireless sensor network, the external environmental factors pose a great challenge on the quality of data collected by a sensor node. Besides, the accuracy and reliability of collected data are also affected by internal factors of a sensor node and the performance of the whole wireless sensor network. Common internal factors are resource limitation such as battery capacity, computing ability, and memory size. Moreover, a wireless sensor network deployed for production use is very likely to suffer from malicious attacks from both inside and outside. Therefore, anomaly is inevitable for time series collected by a wireless sensor network. In the paper, we make an extensive review about two types of anomaly detection methods in time series for wireless sensor networks: outlier detection and pattern anomaly detection. For outlier detection, six categories of detection methods are elaborated. For pattern anomaly detection, two main ideas are discussed. Finally, we present the main drawbacks of existing approaches to outlier detection and pattern anomaly detection for time series. The possible future research directions are also summarized.
At present, accelerated processing of remote sensing big data has become an important research topic in the field of remote sensing. Remote sensing imageprocessing based on large-scale clusters currently is the mains...
详细信息
ISBN:
(数字)9781728172675
ISBN:
(纸本)9781728172682
At present, accelerated processing of remote sensing big data has become an important research topic in the field of remote sensing. Remote sensing imageprocessing based on large-scale clusters currently is the mainstream. However, how to fully tap the computing power of a single computing node in the cluster has become issues that cannot be ignored in the field of remote sensing imageprocessing. Traditional GPU programming is difficult to develop, the development cycle is long, and the requirements for developers are very high. In order to improve the efficiency of GPU programming and shorten the development cycle of parallel programs, Nvidia, Grary, PGI and CAPS jointly launched a new programming standard-OpenAcc. In this paper, OpenAcc-NDVI, as a fast parallel extraction method is used to optimize NDVI algorithm. Based on different computing scenarios, two granularity acceleration models are proposed. It has been verified by multiple experiments that when the data size reaches 10000 * 10000, OpenAcc-NDVI can achieve an acceleration of about 5.3 times. After error analysis, the error of algorithm experiment result is 0, there is no loss of precision. The NDVI algorithm based on OpenAcc has excellent acceleration performance, efficient development process, and high calculation accuracy.
Heart diseases rank among the leading causes of global mortality, demonstrating a crucial need for early diagnosis and intervention. Most traditional electrocardiogram (ECG) based automated diagnosis methods are train...
详细信息
Air pollution is the result of comprehensive evolution of a dynamic and complex system composed of emission sources, topography, meteorology and other environmental factors. The establishment of spatiotemporal evoluti...
详细信息
Thermal imaging is becoming popular recently due to its all-weather and all-time capability. However, video saliency detection, an active topic in computer vision, has not been well studied for thermal videos. In this...
详细信息
This paper develops an end-to-end ECG signal classification algorithm based on a novel segmentation strategy and 1D Convolutional Neural Networks (CNN) to aid the classification of ECG signals and alleviate the worklo...
详细信息
This paper develops an end-to-end ECG signal classification algorithm based on a novel segmentation strategy and 1D Convolutional Neural Networks (CNN) to aid the classification of ECG signals and alleviate the workload of physicians. The ECG segmentation strategy named R-R-R strategy (i.e., retaining ECG data between the R peaks just before and after the current R peak) is used for segmenting the original ECG data into segments to train and test the 1D CNN models. The novel strategy mimics physicians in scanning ECG to a greater extent, and maximizes the inherent information of ECG segments for diagnosis. The performance of the proposed end to end ECG signal classification algorithm was verified with the ECG signals from 48 records in the MIT-BIH arrhythmia database. When the heartbeat types were divided into the five classes recommended by clinicians, i.e., normal beat, left bundle branch block beat, right bundle branch block beat, premature ventricular contraction, and paced beat, the classification accuracy, the area under the curve (AUC), the sensitivity, and the F1-score achieved by the proposed model were 0.9924, 0.9994, 0.99 and 0.99, respectively. When the heartbeat types were divided into six classes recommended by clinicians, i.e., normal beat, left bundle branch block beat, right bundle branch block beat, premature ventricular contraction, paced beat and other beats, the beat classification accuracy, the AUC, the sensitivity, and the F1-score achieved by the model reached 0.9702, 0.9966, 0.97, and 0.97, respectively. When the heartbeat types were divided into five classes recommended by the Association for Advancement of Medical Instrumentation (AAMI), i.e., normal beat, supraventricular ectopic beat, ventricular ectopic beat, fusion beat, and unknown beat, the beat classification accuracy, the sensitivity, and the F1-score were 0.9745, 0.97, and 0.97, respectively. Experimental results show that the proposed method achieves better performance than the s
In large-scale SDN network, a single centralized controller can not meet the demand, and multiple controllers are needed to deal with the problem, which leads to the problem of multi control balanced deployment. In th...
详细信息
Currently, major security incidents caused by the “unlicensed flying” of Unmanned Aerial Vehicle (UAV) have emerged one after another, which poses a grave threat to the security issues of public facilities and sensi...
详细信息
Currently, major security incidents caused by the “unlicensed flying” of Unmanned Aerial Vehicle (UAV) have emerged one after another, which poses a grave threat to the security issues of public facilities and sensitive areas. Whether it can timely detect and prevent “unlicensed flying” of UAV has become a social concern. In response to this demand, the transfer learning method is adopted in this paper to conduct twoclassification and detection on UAV images. image recognition technology based on transfer learning is an effective method to improve recognition accuracy by applying deep learning models to small samples. Different from the large number of training samples required by deep learning, transfer learning transfers the weights of the pre-trained deep neural network, and uses only small sample data to obtain good results in UAV image recognition. First of all, this paper proposes to construct a UAV data set according to different types of UAV shape structures, to perfect the classification and detection effect and the generalization ability of the model. Then, based on the transfer learning method, experimental comparison is made between three classic deep convolutional neural network classification models (Inception V3, ResNet 101 and VGG16) and two classic deep convolutional neural network detection models (Faster RCNN and SSD). Finally, an experimental evaluation is conducted on the collected UAV test data set. Compared with the traditional recognition model, the image classification model based on transfer learning employed in this paper has achieved important improvements in accuracy, recall and precision. Especially in the InceptionV3 model of transfer training, the recall reaches 96.98%. In addition, the image detection model based on transfer learning has achieved good detection results in accuracy, recall and F1-score.
暂无评论