Super-Resolution (SR) techniques for image restoration have recently been gaining attention due to their excellent performance. For powerful learning abilities, Generative Adversarial Networks (GANs) have been proven to have achieved great success. In this paper, we propose an Enhanced Generative Adversarial Network (EGAN) for improving its effects for a real-time Super-Resolution task. The main content of this paper are as follows: (1) We adopted the Laplacian pyramid framework as a pre-trained module, which is beneficial for providing multiscale features for our input. (2) At each feature block, a convolutional skip-connections network, which may contain some latent information, was significant for the generative model to reconstruct a plausible-looking image. (3) Considering that the edge details usually play an important role in image generation, a perceptual loss function was defined to train and seek the optimal parameters. Quantitative and qualitative evaluations were demonstrated so that our algorithm not only took full advantage of the Convolutional Neural Networks (CNNs) to improve the image quality, but also performed better than other algorithms in speed and performance for real-time Super-Resolution tasks.

In order to automate defect detection with few samples using unsupervised learning, this paper, considering materials commonly used in aircraft, proposes a phased array ultrasonic detection defect identification method using non-defect samples for training, and three-dimensional characterization is completed on this basis. A phased array ultrasonic device was used to detect two typical structures: a carbon fiber composite cylinder structure and a metal L-shaped structure. No damage label image was required, and the non-damaged sample was used as the the network training input. Based on contrast learning and the cross-registration loss of common features, a feature-matching network was constructed to extract the common features of undamaged detection data, and the performance was optimized by combining STN and GCNet modules. When the detection data of the sample were input to the aforementioned network, the defect distribution representing the location and rough shape of the defect was obtained through Mahalanobis distance calculation. The length was estimated using the S-scan image sequence sampling method. Additionally, the depth of the hole was estimated by combining the B-scan data with line recognition. According to the original model of the sample, the 3D characterization of defects was completed by pyautocad. In the experimental stage, three ablation experiments were carried out to verify the necessity of each module, and performance comparisons were mainly evaluated by F1 score and visualization using four existing well-known anomaly detection methods.

Spindle stiffness is one of the most critical indicators for evaluating and measuring the service performance of spindles. The traditional static stiffness indexes only involve static analysis and rarely focus on the study of spindle-carrying capacity under operating conditions. In this paper, the explicit solution approach is used to develop a mechanical model of the spindle’s axial operating stiffness. This model was then used to explore the influence of rotational speed on the softening and hardening features of the spindle axial operating stiffness, and experimental verification was carried out. According to studies, the speed of a fixed-position preload spindle can lead its operating stiffness to exhibit a “stiffness-hardening” feature. However, when the axial displacement of the spindle is small, the operating stiffness curve of the spindle displays a noticeable “fluctuation” phenomenon for low-speed spindles. Furthermore, the speed-induced preload has a significant impact on the test results when testing spindle axial operating stiffness.

In this study, A GCN distinction network is proposed to identify the defect status of steel corner structures. In this defect determination method, the detected S-scan data is converted into A-scan by Doppler software to form a two-dimensional detection matrix, and the two dimensions are deflection angle stepping and sampling point on the acoustic distance respectively. GCN needs data node relations for GCN’s adjacent matrix, which concludes the shape of the sample and the reflection information and the structure of acoustic beam direction. The conventional inverse and threshold method are introduced in order to enhance classification performance. According to the node and the adjacency matrix relationship between nodes of affiliation and node sampling value automatically set the node label, The GCN model is constructed. Due to the characteristics of GCN, the beam length in the adjacent matrix cannot be changed again in the application test. Except for the detection model of Beamtool software, all other data processes and distinction are carried out automatically. In this study, the PAUT of low-alloy steel (30CrMnSi) was carried out by the PHASCAN II detector, to verify the effectiveness of the research method through natural wave and defect echo experiments which will be confused to distinguish.

The direct utilization of low-light images hinders downstream visual tasks. Traditional low-light image enhancement (LLIE) methods, such as Retinex-based networks, require image pairs. A spiking-coding methodology called intensity-to-latency has been used to gradually acquire the structural characteristics of an image. convLSTM has been used to connect the features. This study introduces a simplified DCENet to achieve unsupervised LLIE as well as the spiking coding mode of a spiking neural network. It also applies the comprehensive coding features of convLSTM to improve the subjective and objective effects of LLIE. In the ablation experiment for the proposed structure, the convLSTM structure was replaced by a convolutional neural network, and the classical CBAM attention was introduced for comparison. Five objective evaluation metrics were compared with nine LLIE methods that currently exhibit strong comprehensive performance, with PSNR, SSIM, MSE, UQI, and VIFP exceeding the second place at 4.4% (0.8%), 3.9% (17.2%), 0% (15%), 0.1% (0.2%), and 4.3% (0.9%) on the LOL and SCIE datasets. Further experiments of the user study in five non-reference datasets were conducted to subjectively evaluate the effects depicted in the images. These experiments verified the remarkable performance of the proposed method.

The broadening use of carbon-fiber reinforced plastics (CFRP) in various industries has raised safety concerns. To control the hazard, infrared thermography inspection techniques, capable of large-area scanning, instantaneous imaging, and completely non-contact detection, has been increasingly adopted to detect flaws inside CFRP structures. And in practical engineering, post-processing algorithms are required to eliminate noise and enhance defect detectability after initial inspection course. However, rapid updates of infrared camera bring high-resolution thermograms yet. Whereas classic algorithms have encountered data explosion problem which led to low efficiency and engineering infeasibility. Therefore, to promote the efficiency of postprocessing algorithms, we propose equal interval sampling (EIS) method to sample raw thermal data. Two CFRP laminates with embedded artificial Teflon inserts of different sizes at different depths are inspected by a laser infrared thermography system. The obtained raw thermal videos made up of 500 frame thermal images with 512 × 640 pixels were converted to 3D thermographic data. We apply EIS to sample the data by setting different intervals. And the sampled data are fed to classic post-processing algorithms, thermal signal reconstruction (TSR) and principal component analysis (PCA), which have prevailed in processing thermographic data. Regarding different intervals of EIS as variables, the processing efficiency, and the defect detection performance of EIS-TSR and EIS-PCA are studied. The results demonstrate that EIS-TSR method fails to promote processing speed and maintain the performance of defect detection at the same time. In contrast, the EIS-PCA method attains a remarkable efficiency and intact defect detectability simultaneously, manifesting its adequacy in engineering practice.