Recently, research on data-driven bearing fault diagnosis methods has attracted increasing attention due to the availability of massive condition monitoring data. However, most existing methods still have difficulties in learning representative features from the raw data. In addition, they assume that the feature distribution of training data in source domain is the same as that of testing data in target domain, which is invalid in many real-world bearing fault diagnosis problems. Since deep learning has the automatic feature extraction ability and ensemble learning can improve the accuracy and generalization performance of classifiers, this paper proposes a novel bearing fault diagnosis method based on deep convolutional neural network (CNN) and random forest (RF) ensemble learning. Firstly, time domain vibration signals are converted into two dimensional (2D) gray-scale images containing abundant fault information by continuous wavelet transform (CWT). Secondly, a CNN model based on LeNet-5 is built to automatically extract multi-level features that are sensitive to the detection of faults from the images. Finally, the multi-level features containing both local and global information are utilized to diagnose bearing faults by the ensemble of multiple RF classifiers. In particular, low-level features containing local characteristics and accurate details in the hidden layers are combined to improve the diagnostic performance. The effectiveness of the proposed method is validated by two sets of bearing data collected from reliance electric motor and rolling mill, respectively. The experimental results indicate that the proposed method achieves high accuracy in bearing fault diagnosis under complex operational conditions and is superior to traditional methods and standard deep learning methods.

Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them. However, deep learning-based methods still face two challenging problems in practical MI EEG signal classification applications: (1) Generally, training a deep learning model successfully needs a large amount of labeled data. However, most of the EEG signal data is unlabeled and it is quite difficult or even impossible for human experts to label all the signal samples manually. (2) It is extremely time-consuming and computationally expensive to train a deep learning model from scratch. To cope with these two challenges, a deep transfer convolutional neural network (CNN) framework based on VGG-16 is proposed for EEG signal classification. The proposed framework consists of a VGG-16 CNN model pre-trained on the ImageNet and a target CNN model which shares the same structure with VGG-16 except for the softmax output layer. The parameters of the pre-trained VGG-16 CNN model are directly transferred to the target CNN model used for MI EEG signal classification. Then, front-layers parameters in the target model are frozen, while later-layers parameters are fine-tuned by the target MI dataset. The target dataset is composed of time-frequency spectrum images of EEG signals. The performance of the proposed framework is verified on the public benchmark dataset 2b from the BCI competition IV. The experimental results show that the proposed framework improves the accuracy and efficiency performance of EEG signal classification compared with traditional methods, including support vector machine (SVM), artificial neural network (ANN), and standard CNN.

Electroencephalography (EEG) plays an import role in monitoring the brain activities of patients with epilepsy and has been extensively used to diagnose epilepsy. Clinically reading tens or even hundreds of hours of EEG recordings is very time consuming. Therefore, automatic detection of seizure is of great importance. But the huge diversity of EEG signals belonging to different patients makes the task of seizure detection much challenging, for both human experts and automation methods. We propose three deep transfer convolutional neural networks (CNN) for automatic cross-subject seizure detection, based on VGG16, VGG19, and ResNet50, respectively. The original dataset is the CHB-MIT scalp EEG dataset. We use short time Fourier transform to generate time-frequency spectrum images as the input dataset, while positive samples are augmented due to the infrequent nature of seizure. The model parameters pretrained on ImageNet are transferred to our models. And the fine-tuned top layers, with an output layer of two neurons for binary classification (seizure or nonseizure), are trained from scratch. Then, the input dataset are randomly shuffled and divided into three partitions for training, validating, and testing the deep transfer CNNs, respectively. The average accuracies achieved by the deep transfer CNNs based on VGG16, VGG19, and ResNet50 are 97.75%, 98.26%, and 96.17% correspondingly. On those results of experiments, our method could prove to be an effective method for cross-subject seizure detection.

Anomaly detection is an important issue in trajectory data mining. Various approaches have been proposed to address this issue. However, most previous studies focus only on outlier detection but rarely on pattern mining of anomalous trajectories. Mining patterns of anomalous trajectories can reveal the underlying mechanisms of these outliers. This paper studies four distinct patterns of anomalous trajectories, and proposes a method to detect and classify them. First, we present the difference and intersection set (DIS) distance metric to evaluate the similarity between any two trajectories. Based on this distance, we design an anomaly score function to quantify the differences between different types of anomalous trajectories and normal trajectories. We further propose an anomalous trajectory detection and classification (ATDC) method to find anomalies in different anomalous patterns. Finally, we evaluate the proposed ATDC method through extensive experiments on real cab trajectory data. The results show that the proposed approach outperforms existing methods by a significant margin.

At present, the application of Electroencephalogram (EEG) signal classification to human intention-behavior prediction has become a hot topic in the brain computer interface (BCI) research field. In recent studies, the introduction of convolutional neural networks (CNN) has contributed to substantial improvements in the EEG signal classification performance. However, there is still a key challenge with the existing CNN-based EEG signal classification methods, the accuracy of them is not very satisfying. This is because most of the existing methods only utilize the feature maps in the last layer of CNN for EEG signal classification, which might miss some local and detailed information for accurate classification. To address this challenge, this paper proposes a multi-scale CNN model-based EEG signal classification method. In this method, first, the EEG signals are preprocessed and converted to time-frequency images using the short-time Fourier Transform (STFT) technique. Then, a multi-scale CNN model is designed for EEG signal classification, which takes the converted time-frequency image as the input. Especially, in the designed multi-scale CNN model, both the local and global information is taken into consideration. The performance of the proposed method is verified on the benchmark data set 2b used in the BCI contest IV.

As one of the most important artificial intelligence-enabled industrial applications, fault diagnosis is vital in the safe, stable and reliable operation of the equipment. Many existing deep learning-based fault diagnosis methods assume that the distribution of training data is the same as that of testing data, which is almost impossible in practical industrial applications. In addition, most of these fault diagnosis methods are generally memory intensive and computationally expensive. A compressed unsupervised deep domain adaption model-based fault diagnosis method is proposed to overcome the above-mentioned two issues. Firstly, a standard unsupervised domain adaption model is designed to extract the features of training data and testing data, respectively. Then, the maximum mean discrepancy term is introduced to minimize the discrepancy between the extracted features of them. Next, the standard model is compressed through iteratively pruning the redundant convolutional channels. Finally, the obtained compressed model is applied to diagnose faults. The performance of the proposed method is verified on the Case Western Reserve University bearing dataset. Experimental results show that the compressed model can significantly reduce the memory occupation, computational cost and inference time compared with the standard model, but still achieve comparable or even better accuracy on ten transfer diagnostic tasks.