In the study of brain-computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEPs), how to improve classification accuracies BCIs has always been focus researchers. Canonical correlation analysis (CCA) is widely used in BCI systems SSVEPs because its rapidity and scalability. However, classical CCA algorithm encounters difficulty low accuracy a short time. For targetless stimuli, this paper proposes fusion (CCA-CWT-SVM) that combined with CCA, continuous wavelet...

For decades, a great deal of interest in investigating brain network functional connective features has arisen brain-computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEPs). Traditional decoding algorithms, for example, canonical correlation analysis (CCA), only consider the inherent properties each channel terms feature extraction single electroencephalogram (EEG) signal, with inadequate that cannot fully utilize information transmitted by brain. This paper...

In the study of motor imagery (MI) brain-computer interfaces (BCIs), how to improve task classification accuracy has been always one major challenges in applications MI-BCIs. As a type crucial temporal and spatial feature, nonlinear Granger Causality (NGC) analysis was applied feature extraction MI-electroencephalogram (EEG) signals because constructed brain network features can reflect causal relationship between different channels various regions. However, MI-BCI recognition often suffer...

There are massive amounts of civil aviation safety oversight reports collected each year in the China. The narrative texts these typically short texts, recording abnormal events detected during process. In construction an intelligent system, automatic classification is a key and fundamental task. However, all currently analyzed classified into categories by manual work, which time consuming labor intensive. recent years, pre-trained language models have been applied to various text mining...

Background The development of Brain-Computer Interface (BCI) technology has brought tremendous potential to various fields. In recent years, prominent research focused on enhancing the accuracy BCI decoding algorithms by effectively utilizing meaningful features extracted from electroencephalographic (EEG) signals. Objective This paper proposes a method for extracting brain functional network based directed transfer function (DTF) and graph theory. incorporates with common spatial pattern...

Abstract For the brain-computer interface (BCI) system based on steady-state visual evoked potential (SSVEP), it is difficult to obtain satisfactory classification performance for short-time window SSVEP signals by traditional methods. In this paper, a fused multi-subfrequency bands and convolutional block attention module (CBAM) method neural network (CBAM-CNN) proposed discerning SSVEP-BCI tasks. This extracts as initial input of model, then carries out feature fusion all inputs. addition,...

MP3Stego is a typical steganographic tool of MP3 audios. Though many researchers have been making every effort on attacking it, the performance their approaches could be improved especially at low embedding rate. In this paper, we proposed scheme for detecting rate MP3Stego. Based investigating principle and observing alteration quantized MDCT coefficients (QMDCTs), one-step transition probabilities difference were extracted. Finally, SVM was used constructing classification model according...

Aiming at the feature extraction of left- and right-hand movement imagination EEG signals, this paper proposes a multichannel correlation analysis method employs Directed Transfer Function (DTF) to identify connectivity between different channels construct brain network, extract characteristics network information flow. Since flow identified by DTF can also reflect indirect signal networks, newly extracted features are incorporated into traditional AR model parameter extend scope sets....

When a brain-computer interface (BCI) is designed, high classification accuracy difficult to obtain for motor imagery (MI) electroencephalogram (EEG) signals in view of their relatively low signal-to-noise ratio. In this paper, fused multidimensional method based on extreme tree feature selection (FMCM-ETFS) proposed discerning EEG tasks. First, the signal was filtered by Butterworth filter preprocessing. Second, C3, C4, and CZ channels were selected extract time-frequency domain spatial...

The MPEG-1 Audio Layer 3 can be recorded as archive and lawful evidence. However, this MP3 audio may often forged by forgers for their own benefits in some significant events, which will cause double compression. In paper, the statistical features based on scale factors under long window application iterative loop are extracted, a Support Vector Machine is applied classification to detect Experimental results demonstrate that proposed method accurate effective compression detection at...

MP3 is the most widely used audio format nowadays in our daily life and music on internet are often of this format. In order to seek commercial benefit, low bit rate audios usually transcoded high rate, resulting fake-quality audios. Through analyzing frequency statistics Huffman table indexes, a method for detecting such presented paper. Experimental results show effectiveness proposed method.

For the purpose of addressing issue low classification accuracy resulting from signal-to-noise ratios in electroencephalogram (EEG) signals, this paper, a algorithm for motion imagery EEG signals (PSD-PSO-SVM) utilizing power spectral density analysis (PSD) combined with particle swarm optimization method (PSO) improved support vector machine (SVM) is proposed. A first step to extract features signal frequency domain by PSD, and energy densities 0-30Hz delta, theta, alpha, beta frequencies...

Abstract For the brain-computer interface (BCI) system based on steady-state visual evoked potential (SSVEP), it is difficult to obtain satisfactory classification performance for short-time window SSVEP signals by traditional methods. In this paper, a fused multi-subfrequency bands and convolutional block attention module (CBAM) method neural network (CBAM-CNN) proposed discerning SSVEP-BCI tasks. This extracts as initial input of model, then carries out feature fusion all inputs. addition,...