Deep learning-based surface defect detectors play a crucial role in ensuring product quality during inspection processes. However, accurately and efficiently detecting defects remains challenging due to specific characteristics inherent in defective images, involving a high degree of foreground-background similarity, scale variation, and shape variation. To address this challenge, we propose an efficient Transformer-based detection network, ETDNet, consisting of three novel designs to achieve superior performance. Firstly, ETDNet takes a lightweight Vision Transformer to extract representative global features. This approach ensures an accurate feature characterization of defects even with similar backgrounds. Secondly, a channel-modulated feature pyramid network (CM-FPN) is devised to fuse multi-level features and maintain critical information from corresponding levels. Lastly, a novel task-oriented decoupled (TOD) head is introduced to tackle inconsistent representation between classification and regression tasks. The TOD head employs a local feature representation module to learn object-aware local features and introduces a global feature representation module, based on the attention mechanism, to learn content-aware global features. By integrating these two modules into the head, ETDNet can effectively classify and perceive defects with varying shapes and scales. Extensive experiments on various defect detection datasets demonstrate the effectiveness of the proposed ETDNet. For instance, it achieves AP 46.7% (v.s. 45.9%), and AP ₅₀ 80.2% (v.s. 79.1%) with 49 FPS on NeU-DET. Code is available at https://github.com/zht8506/ETDNet.

Underwater image quality assessment (UIQA) plays a crucial role in monitoring and detecting the quality of acquired underwater images in underwater imaging systems. Currently, the investigation of UIQA encounters two major challenges. First, a lack of large-scale UIQA databases for benchmarking UIQA algorithms remains, which greatly restricts the development of UIQA research. The other limitation is that there is a shortage of effective UIQA methods that can faithfully predict underwater image quality. To alleviate these two challenges, in this paper, we first construct a large-scale UIQA database (UIQD). Specifically, UIQD contains a total of 5369 authentic underwater images that span abundant underwater scenes and typical quality degradation conditions. Extensive subjective experiments are executed to annotate the perceived quality of the underwater images in UIQD. Based on an in-depth analysis of underwater image characteristics, we further establish a novel baseline UIQA metric that integrates channel and spatial attention mechanisms and a transformer. Channel- and spatial attention modules are used to capture the image channel and local quality degradations, while the transformer module characterizes the image quality from a global perspective. Multilayer perception is employed to fuse the local and global feature representations and yield the image quality score. Extensive experiments conducted on UIQD demonstrate that the proposed UIQA model achieves superior prediction performance compared with the state-of-the-art UIQA and IQA methods. The proposed UIQD and UIQA models will be released at https://github.com/YT2015?tab=repositories.

The non-intrusive polynomial chaos (NIPC) expansion method is used to quantify the uncertainty of a stochastic system. It potentially reduces the number of numerical simulations in modelling process, thus improving efficiency, whilst ensuring accuracy. However, the number of polynomial bases grows substantially with the increase of random parameters, which may render the technique ineffective due to the excessive computational resources. To address such problems, methods based on the sparse strategy such as the least angle regression (LARS) method with hyperbolic index sets can be used. This paper presents the first work to improve the accuracy of the original LARS method for uncertainty quantification (UQ). We propose an adaptive LARS method in order to quantify the uncertainty of the results from the numerical simulations with higher accuracy than the original LARS method. The proposed method outperforms the original LARS method in terms of accuracy and stability. The L2 regularisation scheme further reduces the number of input samples while maintaining the accuracy of the LARS method.

Deep learning (DL) approaches in underwater acoustic localization (UAL) have gained a great deal of popularity. While numerous works are devoted to improving the localization precision, they neglect another critical challenge inherent in the DL-based UAL problem, i.e., the model's practicality. Advanced DL models generally exhibit extremely high complexity, requiring a large amount of computational resources and resulting in slow inference time. Unfortunately, the limited processing power and real-time demands in oceanic applications make the deployment of complex DL models exceedingly challenging. To address this challenge, this article proposes a lightweight UAL framework based on knowledge distillation (KD) techniques, which effectively reduces the size of a deep UAL model while maintaining competitive performance. Specifically, a dedicated teacher network is designed using attention mechanisms and convolutional neural networks (CNNs). Then, the KD is performed to distill the knowledge from the teacher network into a lightweight student model, such as a three-layer CNN. In practical deployment, only the lightweight student model will be utilized. With the proposed lightweight framework, the student model has 98.68% fewer model parameters and is 87.4% faster in inference time compared to the teacher network, while the prediction accuracy drops to only 1.07% (97.55% $\rightarrow$ 96.48%). In addition, the generalization ability of the student model is examined through transfer learning, where the model is transferred between two different ocean environments. The student model demonstrates a stronger generalization ability compared to the model without the KD process, as it can quickly adapt itself to a new application environment using just 10% of the data.

Existing no-reference (NR) image quality assessment (IQA) metrics are still not convincing for evaluating the quality of the camera-captured images. Toward tackling this issue, we, in this article, establish a novel NR quality metric for quantifying the quality of the camera-captured images reliably. Since the image quality is hierarchically perceived from the low-level preliminary visual perception to the high-level semantic comprehension in the human brain, in our proposed metric, we characterize the image quality by exploiting both the low-level image properties and the high-level semantics of the image. Specifically, we extract a series of low-level features to characterize the fundamental image properties, including the brightness, saturation, contrast, noiseness, sharpness, and naturalness, which are highly indicative of the camera-captured image quality. Correspondingly, the high-level features are designed to characterize the semantics of the image. The low-level and high-level perceptual features play complementary roles in measuring the image quality. To infer the image quality, we employ the support vector regression (SVR) to map all the informative features to a single quality score. Thorough tests conducted on two standard camera-captured image databases demonstrate the effectiveness of the proposed quality metric in assessing the image quality and its superiority over the state-of-the-art NR quality metrics. The source code of the proposed metric for camera-captured images is released at https://github.com/YT2015?tab=repositories .

The detection head constitutes a pivotal component within object detectors, tasked with executing both classification and localization functions. Regrettably, the commonly used parallel head often lacks omni perceptual capabilities, such as deformation perception (DP), global perception (GP), and cross-task perception (CTP). Despite numerous methods attempting to enhance these abilities from a single aspect, achieving a comprehensive and unified solution remains a significant challenge. In response to this challenge, we develop an innovative detection head, termed UniHead, to unify three perceptual abilities simultaneously. More precisely, our approach: 1) introduces DP, enabling the model to adaptively sample object features; 2) proposes a dual-axial aggregation transformer (DAT) to adeptly model long-range dependencies, thereby achieving GP; and 3) devises a cross-task interaction transformer (CIT) that facilitates interaction between the classification and localization branches, thus aligning the two tasks. As a plug-and-play method, the proposed UniHead can be conveniently integrated with existing detectors. Extensive experiments on the COCO dataset demonstrate that our UniHead can bring significant improvements to many detectors. For instance, the UniHead can obtain $+$ 2.7 AP gains in RetinaNet, $+$ 2.9 AP gains in FreeAnchor, and $+$ 2.1 AP gains in GFL. The code is available at https://github.com/zht8506/UniHead.