As an image processing method, underwater image enhancement (UIE) plays an important role in the field of underwater resource detection and engineering research. Currently, the convolutional neural network (CNN)- and Transformer-based methods are the mainstream methods for UIE. However, CNNs usually use pooling to expand the receptive field, which may lead to information loss that is not conducive to feature extraction and analysis. At the same time, edge blurring can easily occur in enhanced images obtained by the existing methods. To address this issue, this paper proposes a framework that combines CNN and Transformer, employs the wavelet transform and inverse wavelet transform for encoding and decoding, and progressively embeds the edge information on the raw image in the encoding process. Specifically, first, features of the raw image and its edge detection image are extracted step by step using the convolution module and the residual dense attention module, respectively, to obtain mixed feature maps of different resolutions. Next, the residual structure Swin Transformer group is used to extract global features. Then, the resulting feature map and the encoder’s hybrid feature map are used for high-resolution feature map reconstruction by the decoder. The experimental results show that the proposed method can achieve an excellent effect in edge information protection and visual reconstruction of images. In addition, the effectiveness of each component of the proposed model is verified by ablation experiments.
To enhance the efficiency of antenna optimization, surrogate model methods can usually be used to replace the full-wave electromagnetic simulation software. Broad learning system (BLS), as an emerging network with strong extraction ability and remarkable computational efficiency, has revolutionized the conventional artificial intelligence (AI) methods and overcome the shortcoming of excessive time-consuming training process in deep learning (DL). However, it is difficult to model the regression relationship between input and output variables in the electromagnetic field with the unsatisfactory fitting capability of the original BLS. In order to further improve the performance of the model and speed up the design of microwave components to achieve more accurate prediction of hard-to-measure quality variables through easy-to-measure parameter variables, the conception of auto-context (AC) for the regression scenario is proposed in this paper, using the current BLS training results as the prior knowledge, which are taken as the context information and combined with the original inputs as new inputs for further training. Based on the previous prediction results, AC learns an iterated low-level and context model and then iterates to approach the ground truth, which is very general and easy to implement. Three antenna examples, including rectangular microstrip antenna (RMSA), circular MSA (CMSA), and printed dipole antenna (PDA), and 10 UCI regression datasets are employed to verify the effectiveness of the proposed model.
Electromagnetic simulation software has become an important tool for antenna design. However, high-fidelity simulation of wideband or ultra-wideband antennas is very expensive. Therefore, antenna optimization design by using an electromagnetic solver may be limited due to its high computational cost. This problem can be alleviated by the utilization of fast and accurate surrogate models. Unfortunately, conventional surrogate models for antenna design are usually prohibitive because training data acquisition is time-consuming. In order to solve the problem, a modeling method named progressive Gaussian process (PGP) is proposed in this study. Specially, when a Gaussian process (GP) is trained, test sample with the largest predictive variance is inputted into an electromagnetic solver to simulate its results. After that, the test sample is added to the training set to train the GP progressively. The process can incrementally increase some important trusted training data and improve the model generalization performance. Based on the proposed PGP, two monopole antennas are optimized. The optimization results show effectiveness and efficiency of the method.
Convolutional neural networks (CNN) have a strong feature extraction ability for images and present a high level of efficiency and accuracy in object detection and image recognition. When CNN is used to model microwave devices, the existing literature generally uses its size parameters as one‐dimensional (1‐D) input, which does not give full play to the image‐processing ability of CNN. In order to make full use of the characteristics of CNN, this letter converts the 1‐D input of microwave devices into the form of an image model, that is, the 1‐D input is transformed into a two‐dimensional (2‐D) matrix composed of 0 and 1 as the input. The image model is combined with CNN, called image‐based CNN (ICNN), which establishes a deep learning surrogate model between the physical parameters and electrical properties of microwave devices and improves the accuracy and generalization ability of the model. Taking the resonant frequency of the microstrip antenna as a simulation example, modelling was carried out by the proposed ICNN and compared with the mainstream machine learning methods. The results show that the proposed method has high convergence and fitting accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.