Electroencephalography (EEG) datasets are characterized by low signal-to-noise signals and unquantifiable noisy labels, which hinder the classification performance in rapid serial visual presentation (RSVP) tasks. Previous approaches primarily relied on supervised learning (SL), which may result in overfitting and reduced generalization performance. In this paper, we propose a novel multi-task collaborative network (MTCN) that integrates both SL and self-supervised learning (SSL) to extract more generalized EEG representations. The original SL task, i.e., the RSVP EEG classification task, is used to capture initial representations and establish classification thresholds for targets and non-targets. Two SSL tasks, including the masked temporal/spatial recognition task, are designed to enhance temporal dynamics extraction and capture the inherent spatial relationships among brain regions, respectively. The MTCN simultaneously learns from multiple tasks to derive a comprehensive representation that captures the essence of all tasks, thus mitigating the risk of overfitting and enhancing generalization performance. Moreover, to facilitate collaboration between SL and SSL, MTCN explicitly decomposes features into task-specific features and task-shared features, leveraging both label information with SL and feature information with SSL. Experiments conducted on THU, CAS, and GIST datasets illustrate the significant advantages of learning more generalized features in RSVP tasks. Our code is publicly accessible at https://github.com/Tammie-Li/MTCN .

In sensitive scenarios, such as meetings, negotiations, and team sports, messages must be conveyed without detection by noncollaborators. Previous methods, such as encrypting messages, eye contact, and micro-gestures, had problems with either inaccurate information transmission or leakage of interaction intentions. To this end, a novel gesture-free hand intention recognition scheme was proposed, that adopted surface electromyography (sEMG) and isometric contraction theory to recognize hand intentions without any gesture. Specifically, this work includes four aspects: first, the experimental system, consisting of the self-conducted myoelectric wristband, the matched host computer software, and the sports platform, is built to get sEMG signals and simulate multiple usage scenarios; second, the paradigm is designed to standard prompt and collect the gesture-free sEMG datasets. Eight-channel signals of ten subjects were recorded twice per subject at about 5–10 days intervals; third, the toolbox integrates preprocessing methods (data segmentation, filter, normalization, etc.), widely used sEMG classification methods, and various plotting functions, to facilitate future research based this dataset; fourth, the benchmark results of widely used methods are provided. The results involve single-day, cross-day, and cross-subject experiments of six-class and 12-class gesture-free hand intention when subjects have different time windows.

The primary challenge facing the field of magnetic anomaly detection (MAD) currently lies in how to effectively improve detection performance in low signal-to-noise ratio (SNR) and real 1/fα noise scenarios. To overcome these difficulties, this article proposes an optimized MAD method based on random forest (RF) classifier. This method utilizes an orthonormal basis function (OBF) detector based on Karhunen-Loève expansion (KLE) to extract energy from the raw data as time domain features. Meanwhile, the spectral information derived from lowpass filtering (LPF) and fast Fourier transform (FFT) serves as frequency domain features of the raw data. The cutoff frequency of the LPF is determined based on a frequency characteristic function that defines the high-frequency boundary of the target signal. Combining these time and frequency domain features, a simulated dataset is constructed for the training and testing of the detection model. Subsequently, the trained model undergoes further validation and evaluation on semi-real and real datasets built upon measured data from a tunnelling magnetoresistance (TMR) magnetic sensor. Through a series of simulations, we demonstrate that our designed method exhibits superior detection capability and stronger generalization ability compared to other similar OBF-based methods. Furthermore, the superiority of this method is also confirmed by experimental results based on actually measured data.

In scenarios where collaborators and non-collaborators coexist, it is important to enable covert information exchange among collaborators while evading detection. Based on isometric muscle contraction, a surface electromyography (sEMG) based gesture-free hand intention recognition system has been developed to transmit messages securely. However, non-stationary sEMG signals, electrode displacement, and individual differences restrict the performance of single-day, cross-day, and cross-subject experiments. To overcome the difficulty, a discrete wavelet transform-based sEMG data alignment (DWT-DA) method is proposed to reduce the distribution difference between the source and target domains. The DWT-DA consists of three stages: time-frequency domain signal decomposition (TFD-SD), subject sub-signal recombination (SR), and time-frequency domain signal reconstruction (TFD-SR). The TFD-SD decomposes the original signal into several sub-signals focused on a specific frequency band or time-frequency region. The SR recombines the decomposed sub-signals to promote complementary enhancement between the features of each sub-signal. The TFD-SR reconstructs the processed sub-signal into a time-domain signal to achieve data augmentation and alignment. Experiments are conducted on a self-collected gesture-free hand intention recognition dataset, demonstrating excellent performance of the method on single-day, cross-day, and cross-subject evaluations. The code is available at https://github.com/ylfzero/EMG_DWT_DA.