Wyniki wyszukiwania - Biblioteka Nauki

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Deep networks for image super-resolution using hierarchical features

100%

Yang X. , Zhang Y. , Zhou D.

tom Vol. 70, nr 1

art. no. e139616

To better extract feature maps from low-resolution (LR) images and recover high-frequency information in the high-resolution (HR) images in image super-resolution (SR), we propose in this paper a new SR algorithm based on a deep convolutional neural network (CNN). The network structure is composed of the feature extraction part and the reconstruction part. The extraction network extracts the feature maps of LR images and uses the sub-pixel convolutional neural network as the up-sampling operator. Skip connection, densely connected neural networks and feature map fusion are used to extract information from hierarchical feature maps at the end of the network, which can effectively reduce the dimension of the feature maps. In the reconstruction network, we add a 3×3 convolution layer based on the original sub-pixel convolution layer, which can allow the reconstruction network to have better nonlinear mapping ability. The experiments show that the algorithm results in a significant improvement in PSNR, SSIM, and human visual effects as compared with some state-of-the-art algorithms based on deep learning.

Camouflage Assessments with Digital Pattern Painting Based on the Multi-Scale Pattern-in-Picture Evaluation Model

75%

Jia Q. , Xu W. , Yi Y. , Liu J. , Hu J. , Zhu L. , Yang X.

tom Vol. 31, no. 2

39--50

In general, traditional evaluations of target camouflage effects are usually conducted based on observational data and general results of statistical analysis. This widely applied methodology quantifies the detection and identification probabilities of camouflage objects but has considerable shortcomings. This data evaluation process is laborious and time-consuming and very low in reproducibility, which sheds light on the necessity of developing a more efficient method in this study field. The growth of computeraided image processing technology provides technical support for camouflage effect evaluation based on digital image processing. Digital pattern painting, which has been previously applied to combat utility uniforms, is a new methodology full of potential due to its broad geographical adaptability. This study proposes a multi-scale pattern-in-picture method to evaluate camouflage effects at different distances. We also established a computer-aided background image library and camouflage assessments with digital simulation and created an evaluation system that could be effectively applied to combat utility uniforms. More than 40 testers participated in this study, who were asked to score the designed camouflage schemes using the evaluation system proposed. The data from simulation assessments and individual evaluations show that the computer-aided simulation assessments conducted as part of this research can efficiently and objectively evaluate the camouflage effect on military objects.