ABSTRACT Vision Transformers (ViTs) have demonstrated strong capabilities in capturing global contextual information compared to convolutional neural networks, making them promising for remote sensing image analysis. However, ViTs often overlook critical local features, limiting their ability to accurately interpret intricate scenes. To address this issue, we propose an adaptive linear hybrid cross attention transformer (ALHCT). It integrates adaptive linear (AL) attention and hybrid cross (HC) attention to simultaneously learn local and global features. AL is introduced into ViT, as it helps reduce computational complexity from exponential to linear scale. Furthermore, ALHCT incorporates two adaptive linear swin transformers (ALST) to achieve multi‐scale feature representation, enabling the model to capture high‐level semantics and fine details. Finally, to enhance global perception and discriminative power, HC attention fuse local and global features which captured by the two ALST. Experiments on three remote sensing datasets demonstrate that ALHCT significantly improves classification accuracy, outperforming several state‐of‐the‐art methods, validating its effectiveness in classifying complex remote sensing scenes.

Focusing on the poor applicability of existing brake stability control methods for intelligent electric vehicles and the problem that the actual braking intention of the driver and the actual running condition of the vehicle are less considered, a layered brake stability control method for electric vehicles is proposed which considers the driver’s braking intention and vehicle state. Firstly, a GRU (Gated Recurrent Unit) neural network with SE (Squeeze Excitation) module mechanism is proposed to obtain the driver’s real braking intention, and a vehicle state recognition algorithm is designed to obtain the real-time longitudinal speed of the vehicle under complex working conditions, which form a closed-loop control structure for the braking system. Secondly, the layered control structure is used to distribute braking force, and the upper control strategy of the braking system with multi-attention mechanism is proposed to obtain the braking torque required for stable braking of the vehicle. Then, the lower level control strategy is used to coordinate the electro-hydraulic braking torque, and the dynamic coordination distribution method of motor braking and hydraulic braking is designed. Finally, the effectiveness and real-time performance of the layered braking stability control method considering driver’s braking intention and vehicle state are verified by joint simulation and real vehicle road experiments. The experiment results show that the slip rate of the proposed braking control method is about 1.5%, the SOC value of the battery increases by 0.14%~0.18%, and the stability coefficient is stable in the range of 0.02~0.04. The braking system control method can not only ensure the braking efficiency and stability of the vehicle, but also effectively recover the braking energy, which provides a new solution for the braking stability control of intelligent vehicles.

Remote Sensing Image Super-Resolution (RSISR) techniques play a crucial role in various remote sensing applications. However, deep learning-based methods applied to RSISR encounter difficulties in learning complex features of remote sensing images and modeling long-term correlations between pixels. This study proposes aMulti-Scale Sliding Window Attention Generation Adversarial Network (MSWAGAN), which combines the advantages of Convolutional Neural Networks (CNN) and Transformers to overcome these limitations. The MSWAGAN consists of three main parts. In the shallow feature extraction part, CNN is used to extract shallow features from remote sensing images. The deep feature extraction part is divided into two stages. Firstly, amulti-scale sliding window attention (MSWA)is designed to replace the multi-head attention (MHA) in the Transformer. MSWA can learn local multi-scale complex features of remote sensing images without increasing the number of parameters in MHA. Then, the Transformer is utilized to learn global image features and model the long-range correlations between pixels. The image reconstruction part utilizes sub-pixel convolution for feature upsampling. Furthermore, in order to extend the application of super-resolution remote sensing images, a cross-sensor real multi-spectral RSISR dataset consisting of Landsat-8 (L8) and Sentinel-2 (S2) images was constructed, and a series of experiments to improve the spatial resolution of L8 images from 30m to 10m in B, G, R and Near Infrared (NIR) bands were conducted. Experimental results demonstrate that our method outperforms some of the latest SR methods.

Focusing on the low efficiency of the current active safety control method with intelligent networked vehicles, which cannot warn potential dangers in advance, a digital twin vehicle stability monitoring system based on side slip angle is proposed. By analyzing the practical significance of digital twin technology in automobile field, the framework of vehicle stability monitoring system is proposed, which includes vehicle physical system, virtual vehicle model system, vehicle twin data platform and comprehensive monitoring system. Firstly, a virtual vehicle model is established, and its accuracy and real-time performance are verified by real vehicle test. The PSOLSTM (Particle Swarm Optimization Long and Short-Term Memory) algorithm relied on the improved LSTM (Long and Short-Term Memory) is designed in the cause of construct automobile side slip angle state estimator model, which has better accuracy and followability. Secondly, a simulation experiment platform and a real vehicle test platform are built based on the comprehensive monitoring system to verify the accuracy and real-time performance of the designed vehicle side slip angle state estimator model. The experiments show that the maximum estimation error of automobile side slip angle is only 0.634deg under four different working conditions. Finally, a digital twin vehicle stability monitoring platform based on side slip angle is designed. The new intelligent vehicle active safety control mode of “data-driven, virtual-real combination, accurate estimation, autonomous decision-making, shared autonomy” is realized under the drive of digital twin.

In the absence of high-resolution sensors, super-resolution( SR) algorithms for remote sensing imagery improve the spatial resolution of the images. Currently, most of the SR algorithms are based on deep learning methods e.g., convolutional neural networks(CNN). Particularly, the generative adversarial networks(GANs) have demonstrated accepted performances in image super-resolution owing to their powerful generative capabilities. However, remote sensing images have complex feature types, which largely limits the performances of GAN-based SR methods for real satellite images. To address this issue, an attention mechanism and a multi-scale structure are introduced into the generator of the GAN network, and a multi-scale attention GAN(MSAGAN) is constructed in this study. We sequentially arrange the channel attention module and the spatial attention module after the multi-scale structure to emphasize important information, suppress unimportant information details, improve the model’s performance. Furthermore, we add residual connections and dense blocks to further enhance the performance of the generative network by increasing its depth. We compared to other existing deep learning-based SR methods, our proposed MSAGAN algorithm performed better in generating high spatial satellite images.

Contrastive self-supervised learning (CSSL) is a promising method in extracting effective features from unlabeled data. It performs well in image-level tasks, such as image classification and retrieval. However, the existing CSSL methods are not suitable for pixel-level tasks, e.g., change detection (CD), since they ignore the correlation between local patches or pixels. In this paper, we firstly propose a multi-cue contrastive self-supervised learning (MC-CSSL) method to derive dense features for change detection. Besides data augmentation, the MC-CSSL takes advantage of more cues based on the semantic meaning and temporal correlation of local patches. Specially, the positive pair is built from local patches with the similar semantic meaning or temporal ones with the same geographic location. The assumption is that local patches belonging to the same kind of land-covering tend to share similar features. Secondly, the affinity matrix is truncated and introduced to extract change information between two temporal patches obtained from different types of sensors. As a result, some initial unchanged pixels are selected to serve as the supervision for mapping the dense features into a consistent space. Based on the distance between all bi-temporal pixels in the consistent space, a difference image (DI) is generated and more unchanged pixels can be available. The dense feature mapping and unchanged pixel updating proceed alternately. The proposed CD method is evaluated in both homogeneous and heterogeneous cases and the experimental results demonstrate its effectiveness and priority after comparison with some existing state-of-the-art methods. The source code will be available at https://github.com/Yang202308/ChangeDetection_CSSL.

Convolutional neural networks (CNNs) have superior feature learning capabilities with large numbers of labeled samples. The reality is that labeling these samples is costly in terms of human labor. Existing data augmentation methods alleviate the scarcity of labeled samples. However, these methods are not suitable for synthetic aperture radar (SAR) images, owing to special imaging mechanisms and observational objects. The generative SAR images by existing augmented methods show structure distortion. To address this issue, we introduce a curvelet adversarial augmented neural network (CA2NN) for SAR image classification. Specifically, an$\text{A}^{2}$NN is established, which consists of two generative streams and one discriminative stream. In the generative stream, through the mutual transformation between the whole and partial images, more new samples with structural consistency are generated to augment the limited labeled data. In the discriminative stream, these generated samples show certain appearance variations after adversarial training based on the novel joint discriminant criterion. Simultaneously, given the multiscale and multidirectional nature of SAR images, we construct discretized curvelet in 2-D space, aiming to extract the singularity features and avoid overfitting. By integrating curvelet kernels into$\text{A}^{2}$NN, CA2NN can automatically generate more representative features adapting to complex terrain, while greatly reducing the complexity of the network. Experiments are conducted on the SAR images with large-scale and complex scenes, suggesting that the proposed approach significantly improves the classification performance with few labeled samples.

Contrastive self-supervised learning (CSSL) has achieved promising results in extracting visual features from unlabeled data. Most of the current CSSL methods are used to learn global image features with low-resolution that are not suitable or efficient for pixel-level tasks. In this paper, we propose a coarse-to-fine CSSL framework based on a novel contrasting strategy to address this problem. It consists of two stages, one for encoder pre-training to learn global features and the other for decoder pre-training to derive local features. Firstly, the novel contrasting strategy takes advantage of the spatial structure and semantic meaning of different regions and provides more cues to learn than that relying only on data augmentation. Specifically, a positive pair is built from two nearby patches sampled along the direction of the texture if they fall into the same cluster. A negative pair is generated from different clusters. When the novel contrasting strategy is applied to the coarse-to-fine CSSL framework, global and local features are learned successively by forcing the positive pair close to each other and the negative pair apart in an embedding space. Secondly, a discriminant constraint is incorporated into the per-pixel classification model to maximize the inter-class distance. It makes the classification model more competent at distinguishing between different categories that have similar appearance. Finally, the proposed method is validated on four SAR images for land-cover classification with limited labeled data and substantially improves the experimental results. The effectiveness of the proposed method is demonstrated in pixel-level tasks after comparison with the state-of-the-art methods.

Missing data imputation is an essential component of a robust traffic surveillance system. Despite the progress of data imputation technologies in intelligent transportation systems (ITS), only limited efforts have been devoted to tackling data missing issues on signalized roads, which present distinct missing patterns due to the effect of traffic signal timing on traffic flow. Unlike most studies that try to impute missing data at the granularity of road segment and aggregated time intervals, we impute missing data for individual traffic lanes at the granularity of signal cycle. We develop a data-driven fine-grained imputation approach based on a novel gated attentional generative adversarial network (GaGAN), which is highly responsive to the dynamic traffic environments of signalized road networks. The advantage of the network lies in that it can automatically learn inter-lane spatio-temporal correlations during each signal cycle. To model the spatial correlations, we jointly leverage spatial attention mechanisms and graph convolutional operations to quantify inter-lane influences within each signal cycle. To model the temporal correlations, we propose to integrate self-attention mechanisms into gated recurrent units (termed as SA-GRU). Two sub-discriminators are developed to model lane-level complete and missing data distributions, respectively, with the goal to improve the consistency between imputed data and overall data distribution. Experimental results demonstrate that the proposed approach is superior to other state-of-the-art methods, achieving robust performance over different data missing types.

This paper proposes a summing node test method for the operational amplifier and shows the followings: (i) It can be used for parallel testing of multiple AC characteristics (such as open loop gain (AOL), PSRR and CMRR) of one operational amplifier simultaneously with the equivalent accuracy but much faster compared to the NULL method. Also it can measure them even for multiple operational amplifiers at the same time. (ii) It can measure THD, SNR and THD+N of the operational amplifier with the comparable accuracy to the audio analyzer usage case, by applying proper analog filters. In other words, it can measure them with remarkable accuracy at very low cost. These have been verified with simulations and experiments. The proposed summing node test method uses an inverting operational amplifier under test and its negative input is amplified by an auxiliary non-inverting operational amplifier. The input and power supply voltages for the operational amplifier under test are modulated by AC signals with different frequencies. The auxiliary amplifier output is digitized after analog filtering and FFT is performed to the digitized data. This proposed method can reduce operational amplifier test time with good accuracy but without expensive instruments at mass production shipping, to meet the requirements for IoT and automotive as well as audio applications.

This article proposes a novel autoencoder (AE) network based on the distribution of polarimetric synthetic aperture radar (POL-SAR) data matrix, called a mixture autoencoder (MAE). Through a detailed analysis of the data distribution POL-SAR data matrix, a normalization method is also presented in succession. The proposed MAE defines the data error term in the loss function according to the data distribution. It can be regarded as a process of unsupervised feature extraction designed specifically for POL-SAR data matrix. Then, a softmax classifier is trained with the help of data features and the corresponding label information. Next, a stacked MAE (SMAE) network is reasonably constructed by considering the data distribution among different layers. Finally, this article also presents a classification network through discarding the decoder process of the proposed SMAE and connecting with a softmax classifier. The SMAE is trained layer by layer using the unlabeled data. The softmax classifier is also trained with a small number of labeled pixels. With parameters obtained from the above-mentioned procedures as the initial parameters, the whole classification network is trained by the labeled pixels to get a well-trained model, which is used for predicting the corresponding label of the pixel in the data set. Three real POL-SAR data sets, including the AIR-SAR L-band data of Flevoland, The Netherlands, are used in the experiments. Compared with one classical algorithm and two related models with the similar structure, both the proposed methods show improvements in overall accuracy and efficiency as well as possess better adaptability of the parameter and preferable consistency with the classification performance.

This paper describes an accurate and fast testing technique for small DC offset voltage of a high precision operational amplifier. Chopper techniques for DC-AC conversion and FFT spectrum analysis are combined and then accurate DC voltage measurement on the order of μV can be achieved. We have also investigated thermo-electromotive force effects and their countermeasures. Their simulations and experiments with prototype measurement systems have been carried out, and the measurement linearity up to as low as 0.2 μV of the DC measurement voltage was confirmed. We have also investigated its extension to multi-channel realization for short testing time.