Driver Behavior and Intention Recognition Based on Wavelet Denoising and Bayesian Theory

Classification of electroencephalogram (EEG) signals is an important task in the brain computer interface system. This paper presents two combination strategies of feature extraction on EEG signals. In the first strategy, Autoregressive coefficients and approximate entropy are calculated respectively, and the features are obtained by assembling them. In the second strategy, the EEG signals are first decomposed into sub-bands by wavelet packet decomposition. Wavelet packet coefficients are then sent to the autoregressive model to calculate autoregressive coefficients, which are used as features extracted from the original EEG signals. These features are fed to support vector machine for classifying the EEG signals. The classification accuracy has been used for evaluating the classification performance. Experimental results in five mental tasks show that the combination strategies can effectively improve the classification performance when the order of autoregressive model is greater than 5, and the second strategy is superior to the first one in terms of the classification accuracy.

Driving intention recognition is an important aspect of Advanced Driving Assistance Systems (ADAS) for giving drivers suggestions to maneuver safely. The intention recognition algorithms in ADAS are often developed using Machine Learning-based models. The model's input, such as environmental (ENV) and eye-tracking (ET) features affect the model's recognition performance. In this contribution, an Artificial Neural Network-based state machine is used for lane changing intention recognition. Three lane changing behaviors are considered, left/right lane change and lane keeping. Here, data consisting of ENV and ET information are collected using a driving simulator and eye-tracker. The aim is to investigate the effect of different feature types on the model's intention recognition performance. First, a 10-cross validation is performed to evaluate the model's performance, using only ENV and both ENV and ET features. The validation results show that the model with only ENV features performs better with respect to different metrics. Thus in the test, only ENV features are used to evaluate the performance. Accuracy values of higher than 80 % are achieved. Furthermore, the recognition performance of the model is compared with other Machine Learning models. The approach introduced outperforms other models in most metrics.

A new approach for automatic detection of focal EEG signals using wavelet packet decomposition and quad binary pattern method

Partial discharge (PD) in insulators will not only lead to the gradual degradation of insulation performance but even cause power system failure in serious cases. Because there is strong noise interference in the field, it is difficult to accurately locate the position of the PD source. Therefore, this paper proposes a three-dimensional spatial localization method of the PD source with a four-element ultra-high-frequency (UHF) array based on improved wavelet packet dynamic threshold denoising and the generalized cross-correlation algorithm. Firstly, considering the field noise interference, the PD signal is decomposed into sub-signals with different frequency bands by the wavelet packet, and the corresponding wavelet packet coefficients are extracted. By using the improved threshold function to process the wavelet packet coefficients, the PD signal with low distortion rate and high signal-to-noise ratio (SNR) is reconstructed. Secondly, in order to solve the problem that the amplitude of the first wave of the PD signal is small and the SNR is low, an improved weighting function, , is proposed, which is based on the self-power spectral density of the signal and is adjusted by introducing an exponential factor to improve the accuracy of the first wave arrival time and time difference calculation. Finally, the influence of different sensor array shapes and PD source positions on the localization results is analyzed, and a reasonable arrangement scheme is found. In order to verify the performance of the proposed method, simulation and experimental analysis are carried out. The results show that the improved wavelet packet denoising algorithm can effectively realize the separation of PD signal and noise and improve the SNR of the localization signal with low distortion rate. The improved weighting function significantly improves the estimation accuracy of the time difference between UHF sensors. With the sensor array designed in this paper, the relative localization error is 3.46%, and the absolute error is within 6 cm, which meets the requirements of engineering applications.

Vehicle trajectory prediction and driving intention recognition are essential technologies for improving safety and optimizing traffic efficiency in autonomous driving systems. Although traditional LSTM models are effective in trajectory prediction, their complexity and computational requirements impede practical implementation efficiency. To overcome this challenge, this study proposes a hybrid model, MTF-GRU, which integrates multi-feature fusion. Initially, the datasets are preprocessed in this study through denoising, feature extraction, and timing extraction to capture vehicle information from single and fused multi-features. Subsequently, a GRU encoding-decoding model is developed. The encoder processes the feature data to generate context vectors, while the decoder employs a combination of recursive and teaching-driven input modes. Furthermore, a teaching rate control mechanism is integrated to dynamically convert context vectors into future trajectories. The proposed model is validated using the NGSIM datasets, demonstrating superior prediction performance with multi-feature inputs outperforming single features by reducing the average endpoint displacement error by 20.5%. Our model also achieves improved accuracy rates, particularly excelling in long-term predictions with an endpoint displacement error of only 2.31 meters at 5 seconds. Moreover, the overall accuracy rate for lane change intention recognition reaches 91.3%. The model's computational efficiency supports practical deployment in real-time autonomous systems, while future efforts will integrate multi-modal sensor data to enhance adaptability in complex urban scenarios and extreme conditions. Further validation will extend to diverse traffic environments and edge computing platforms to optimize real-world robustness.

The vibration signals of diesel include excess noise that must be eliminated before extraction of characteristic parameters. Firstly, the effects of vibration-signal de-noising among Fourier transform, wavelet decomposition and wavelet packet decomposition are compared. Secondly, singular value decomposition is applied to de-noising vibration signals. Finally, a new de-noise method integrated with wavelet packet and singular value is presented. In this method, vibration signals are decomposed by wavelet packet, and the wavelet packet coefficient is de-noised by singular value decomposition again. The results indicate that the new de-noising method is the best. The SNR (signal-to-noise ratio) of the vibration signals of a diesel cylinder lid is the highest. The diesel vibration waveforms of combustion and valve become clear and the extracted characteristic parameters become more precise.

Multi-scale driver behavior modeling based on deep spatial-temporal representation for intelligent vehicles

Recently, Brain-Computer Interface (BCI) technology has been applied more and more in the field of clinical rehabilitation, which provides an effective way of communication for patients with brain disability and stroke. Because electroencephalogram (EEG) signals are extremely complex and contain many redundant signals, the recognition effect of BCI system based on EEG is not very well. The purpose of this paper is to solve the problem of brain intention classification in brain-computer interface system, which combines Deep Separation Convolutional Neural Network (DSCNN) and Gate Recurrent Unit (GRU) network to classify the motor imagination task of EEG. Firstly, one-dimensional timing EEG signals are transformed into two-dimensional array, and the temporal and spatial features of EEG signals are extracted by separate convolution. Then, these EEG signals containing spatio-temporal features are convolved once to extract spatial features, and the temporal features is extracted by GRU. Finally, the experiments shows that the final intention recognition accuracy reach 97.76% via the open physiological motor imagery data set EEGMMIDB, which is superior to some advanced research methods for motor imagery task recognition at present and helpful to restore the rehabilitation ability of patients with brain injury.

In the wavelet packet threshold denoising algorithm, retaining as many of the original signal wavelet packet coefficients as possible is necessary while eliminating noise. Due to the lack of an adaptive parameter, the traditional soft and hard threshold functions cannot be adjusted adaptively, and the useful signal is always mixed with noise, which leads to insecurity of the Internet of Things (IoT) application. Therefore, this paper discusses the significance and feasibility of sample entropy as an adaptive parameter in the wavelet packet threshold function. Furthermore, a new threshold function is proposed to improve the traditional wavelet packet denoising algorithm. The algorithm can process the wavelet packet coefficients differently for different noise backgrounds to achieve better denoising effect, thereby improving the security of the IoT. The simulation results show that compared with the classical wavelet packet threshold function, the proposed improved algorithm achieves better denoising effect and is beneficial to enhance the security of IoT applications.

It was suggested that spectrum estimation can be accomplished by applying wavelet denoising methodology to wavelet packet coefficients derived from the logarithm of a spectrum estimate. The particular algorithm we consider consists of computing the logarithm of the multitaper spectrum estimator, applying an orthonormal transform derived from a wavelet packet tree to the log multitaper spectrum ordinates, thresholding the empirical wavelet packet coefficients, and then inverting the transform. For a small number of tapers, suitable transforms/partitions for the logarithm of the multitaper spectrum estimator are derived using a method matched to statistical thresholding properties. The partitions thus derived starting from different stationary time series are all similar and easily derived, and any differences between the wavelet packet and discrete wavelet transform (DWT) approaches are minimal. For a larger number of tapers, where the chosen parameters satisfy the conditions of a proven theorem, the simple DWT again emerges as appropriate. Hence, using our approach to thresholding and the method of partitioning, we conclude that the DWT approach is a very adequate wavelet-based approach and that the use of wavelet packets is unnecessary.

The research of an effective denoising algorithm for the actual obtained signals with chaotic characteristics is of great interest to all fields of the subject. To show the true chaotic state and analyze the dynamic characteristics of the chaotic system more accurately, an improved denoising algorithm using wavelet packet is proposed in this paper. Wavelet packet decomposition has an optimal sub-band tree structure, which can be used for local analysis of chaotic signals. Based on the correlation function value differences of wavelet packet coefficients, the algorithm determines the optimal decomposition layer, while the optimal wavelet packet basis is obtained with logarithmic energy entropy as the cost function. Furthermore, on the one hand, it makes efforts to divide wavelet packet coefficients into approximate parts, fuzzy parts and detail parts. On the other, it carries out singular spectrum analysis, the fuzzy threshold and the correlation analysis for the select of these three different types of coefficients in order to retain the dynamic performance of chaotic signals in the greatest extent. To verify the effectiveness of the algorithm, the Lorenz chaotic model is employed to analyzed. Simulation results verify the practicability of the improved denoising algorithm, which can also be well applied to various chaotic signals denoising with different noise levels.

For more than 30 years, pedestrian safety studies have considered pedestrian–vehicle collision patterns and pedestrian and driver behavior at marked and unmarked crosswalks at uncontrolled crossings. Recent research in this area, conducted by the University of California, Berkeley, Traffic Safety Center on behalf of the California Department of Transportation (Caltrans), was designed to fill key gaps in the literature by analyzing driver and pedestrian behavior and knowledge of right-of-way laws for marked and unmarked crosswalks. The Caltrans study, as with most previous crosswalk studies, focused on urban and suburban areas (in this case, the San Francisco, California, Bay Area), where driver and pedestrian characteristics do not change significantly from day to day. After this study came the recognition that similar research was needed in rural and recreational locations, where the population frequently changes. As such, this paper summarizes results from field observations of driver and pedestrian behavior at marked and unmarked crosswalks at uncontrolled crossings during the summer in the Tahoe Basin of California. This study, also funded by Caltrans, concludes that the behavior trends identified in the study of urban and suburban areas in the Bay Area are largely similar in a rural and recreational context. This finding is significant for Caltrans, a statewide agency that is seeking to provide a consistent crosswalk installation and treatment policy for its facilities across California. Other regional and state agencies may similarly benefit from the findings of the study.

The necessity of intelligent driver assistance systems in every future car is inevitable with the increased of injuries and fatalities from road accidents. The implementation of a wide array of sensors in a car could be utilized to detect driver's behavior to improve safety. This paper presents an analysis of car driver's behavior pattern recognition in lane changes and turns behavior using context-free grammar in linguistic-based syntactic recognition approach. The driving data which consists of the driver's eye gaze; the car's speed, steering wheel angle, and signal indication; are represented in symbolic form thus forming sentences which represent a specific behavior. Grammars are formulated based on the sentences trained from training data. The grammars are then evaluated by parsing the testing data. The experimental results show that driving behaviors are correctly detected with an average of 84.3% recognition rate exactly before a driving maneuver commences, which gives time to caution the driver should there any danger ahead.

Driver behavior has been considered as the most critical and uncertain criteria in the study of traffic safety issues. Driver behavior identification and categorization by using the Fuzzy Analytic Hierarchy Process (FAHP) can overcome the uncertainty of driver behavior by capturing the ambiguity of driver thinking style. The main goal of this paper is to examine the significant driver behavior criteria that influence traffic safety for different traffic cultures such as Hungary, Turkey, Pakistan and China. The study utilized the FAHP framework to compare and quantify the driver behavior criteria designed on a three-level hierarchical structure. The FAHP procedure computed the weight factors and ranked the significant driver behavior criteria based on pairwise comparisons (PCs) of driver’s responses on the Driver Behavior Questionnaire (DBQ). The study results observed “violations” as the most significant driver behavior criteria for level 1 by all nominated regions except Hungary. While for level 2, “aggressive violations” is observed as the most significant driver behavior criteria by all regions except Turkey. Moreover, for level 3, Hungary and Turkey drivers evaluated the “drive with alcohol use” as the most significant driver behavior criteria. While Pakistan and China drivers evaluated the “fail to yield pedestrian” as the most significant driver behavior criteria. Finally, Kendall’s agreement test was performed to measure the agreement degree between observed groups for each level in a hierarchical structure. The methodology applied can be easily transferable to other study areas and our results in this study can be helpful for the drivers of each region to focus on highlighted significant driver behavior criteria to reduce fatal and seriously injured traffic accidents.

Electroconvulsive therapy (ECT) is an effective treatment for severe depression. In this paper, we have used an algorithm based on wavelet packet (WP) analysis of EEG signals to detect seizures induced by ECT. After determining dominant frequency bands in the ictal period during ECT, the energy ratio of these bands was computed using the corresponding WP coefficients. This ratio was then used as an index to recognize seizure periods. Four different approaches to detect ECT seizures were employed in 41 EEG recordings from nine patients. Sensitivity in ECT seizure detection ranged from 76 to 95% while the false detection rate ranged from 6 to 13.