Carbon fiber reinforced polymer (CFRP) has been increasingly used in aviation industry since it significantly enhances the performance of aircraft. However, imperfections inside CFRP structures pose a threat to safety. Apart from defect shape and size, flaw depth is crucial assess severity. In this work, we utilize laser infrared thermography (LIT) system inspect an sheet adopt long-short term memory recurrent neural network (LSTM-RNN) determine depth. Thermographic sequences obtained by LIT...

Performance evaluation method that can compare and analyze different fusion techniques is an essential part in the development of image techniques. This paper mainly discusses applications restrictions performance metrics such as statistics feature related, information theory correlation entropy, deviation etc. Accurate effective assessment performances may instruct methods, increase efficiency system realization.

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) proven achieved great success. In this paper, we propose an Enhanced Network (EGAN) improving its effects a real-time task. The main content of paper are as follows: (1) We adopted the Laplacian pyramid framework pre-trained module, which is beneficial providing multiscale features our input....

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 identification method non-defect for training, and three-dimensional characterization is completed on basis. A device was detect two typical structures: carbon fiber composite cylinder structure metal L-shaped structure. No damage label image required, the non-damaged sample as network training input. Based...

In this paper, we propose a Terahertz image Super-Resolution (SR) method based on very deep Convolutional Networks (VDCN) in non-destructive testing (NDT) application. Glass fiber reinforced plastic (GFRP) with prefabrication defects is used as test specimen. Through fast scanning imaging by domain spectrometer, obtained the single original image. Then HR can be generated trained VDCN. Experimental results show that restored our has higher peak signal-to-noise ratio (PSNR) and structural...

Image fusion is becoming one of the hottest techniques in image processing. Many methods have been developed a number applications. This paper mainly discusses structures process, which classified as hierarchical structure, overall and arbitrary structure. And effects such on performances are analyzed. In experiment, typical hyperspectral data set fused using same wavelet transform based technique, but applying different structures. The differences among their images experimental results...

Recently, terahertz (THz) nondestructive testing technology has prevailed in the defect detection of composite materials yet is limited by identifying deeper flaws. To solve this problem, a time-domain piecewise imaging method based on signal extraction proposed article. First, surface reflection echo removed applying time correction, and then segmented according to thickness sample. Finally, high-quality images are obtained image fusion technology. All methods, including traditional ones,...

Spindle stiffness is one of the most critical indicators for evaluating and measuring service performance spindles. The traditional static indexes only involve analysis rarely focus on study spindle-carrying capacity under operating conditions. In this paper, explicit solution approach used to develop a mechanical model spindle’s axial stiffness. This was then explore influence rotational speed softening hardening features spindle stiffness, experimental verification carried out. According...

We demonstrate a deep residual network with transfer learning strategy for automatic terahertz (THz) recognition of the hidden defects in fiber reinforced polymer (FRP) composites small-scale training data. The performance high accuracy, precision, sensitivity and specificity indicate effectiveness proposed method automatically identifying different THz nondestructive applications.

The radome composing of GFRP with sandwich structure is easy to generate impact damage, core material shedding and fiber delamination during the service causing structural strength performance waves fall. Then it affects normal operation radar antenna system. characteristics terahertz wave contain low photon energy, strong penetrating ability dielectric material, transient, broadband coherence. simulation experiment reflective NDT (nondestructive testing) overlapped multi-delamination...

In recent years, terahertz technology as a new detection has developed rapidly, but the of carbon fiber composites great limitations. this paper, firstly, electrical anisotropy and skin depth are analyzed, then experimental samples tested by THz time-domain spectroscopy. Finally, results calculation experiment compared analyzed. The show that in transmission mode, ability is related to depth, which further verified theoretical calculation. quantitative explanation for limitation given first...

The fuzzy systems theories were effective methods widely used in modern engineering realm where data uncertain and lacking. Fuzzy grey the double-characters of most information assessment. whitenization incidence models came from supportability assessment, here enlarged normal Accordingly, modified. new model called was initially presented to settle problem lacking a processing method based on basic theory indicates that it is easier confirm relativeness than absoluteness. Based whitening...

Wavelet transform (WT) based image fusion techniques have been widely studied and applied in hyperspectral data processing research. Generally, decomposition, sub-image fusion, reconstruction are the main steps for a WT fusion. The performances of methods basically dependent on scheme its related transform. In this paper, we mainly discuss effects wavelets process. choice wavelet bases approach is also discussed from standpoint computation cost

This paper addresses a new micromechanical model to predict biaxial tensile moduli of plain weave fabric (PWF) composites by considering the interaction between orthogonal interlacing strands. The two yarns in unit cell (UC) were idealized as curved beams with path depicted using sinusoidal shape functions. PWF derived means minimum total complementary potential energy principle founded on micromechanics. tests respectively conducted RTM-made EW220/5284 at five loading ratios 0, 1, 2, 3 and...