An automatic colorization algorithm can convert a grayscale image to a colorful image using regression loss functions or classification loss functions. However, the regression loss function leads to brown results, while the classification loss function leads to the problem of color overflow and the computation of the color categories and balance weights of the ground truth required for the weighted classification loss is too large. In this paper, we propose a new method to compute color categories and balance the weights of color images. In this paper, we propose a new method to compute color categories and balance weights of color images. Furthermore, we propose a U-Net-based colorization network. First, we propose a category conversion module and a category balance module to obtain the color categories and to balance weights, which dramatically reduces the training time. Second, we construct a classification subnetwork to constrain the colorization network with category loss, which improves the colorization accuracy and saturation. Finally, we introduce an asymmetric feature fusion (AFF) module to fuse the multiscale features, which effectively prevents color overflow and improves the colorization effect. The experiments show that our colorization network has peak signal-to-noise ratio (PSNR) and structure similarity index measure (SSIM) metrics of 25.8803 and 0.9368, respectively, for the ImageNet dataset. As compared with existing algorithms, our algorithm produces colorful images with vivid colors, no significant color overflow, and higher saturation.

Compound aircraft, merging the advantages of two or more aircraft, is now a promising concept for designing modern aircraft. How to achieve successful docking of two aircraft in flight is an important issue for the applications of vertical compound aircraft. In order to resolve such an issue, this paper proposes a new visual based navigation docking scheme, which is different from rendezvous and docking operations in space applications. The present scheme uses a monocular camera mounted on the chaser aircraft with a certain installation angle to collect the visual features of the target aircraft, then the position and attitude of the landing gear of the target aircraft can be determined. Moreover, a six degrees of freedom docking mechanism mounted on the chaser aircraft is designed to catch hold of the landing gear tires of the target aircraft. In order to avoid undesired collision during the docking process, Z-direction position of the target landing gear is predicted and the proposed Z-direction speed control algorithm is applied to the chaser platform. The robustness of the present scheme has been validated numerically and experimentally by means of a chaser platform and a target platform on the ground testbed. Successful docking of the chaser platform and the target landing gear has been achieved, respectively, when the target platform performs a Z-direction movement and a compound movement.

More and more work has focused on incorporating different kinds of literals into Knowledge Graph to promote the performance of knowledge embedding. These literals contain numeric literals, text literals, image literals and so on. These additional descriptions are connected to the entities through certain attributes. To incorporate numeric literals, some methods combine the embeddings of literals part with the traditional part - embeddings of entities. However, in the construction of literals embeddings, these existing methods consider the differences of these attributes: one dimension represents one attribute. But they ignore semantic meanings of attributes themselves. In this paper, we propose two methods to incorporate attributes semantics into knowledge graph embeddings from two perspectives: LiteralEAN and literalE-AT. They concatenate with the embeddings of numeric literals by different ways. Furthermore, their extension model LiteralE-C is also proposed as having a more comprehensive representation of attributes semantics. In an empirical study over two standard datasets FB15k and FB15k-237, we evaluate our models for link prediction. We demonstrate that they show an effective way to improve LiteralE and achieve state-of-the-art results. In ablation experiments, we find combined models do better than their singular counterparts in most cases.

In recent years, the wargaming environment has attracted much attention from researchers in the field of intelligent decision technology. Its appeal stems mainly from challenging features such as imperfect information, asymmetric environments, and long-term decision-making in wargames. In this article, we devise a coding framework for a small-scale wargame and develop an intelligent inference system based on the proposed framework. Meanwhile, we conduct intelligent deduction experiments using this system and thoroughly analyze the results. Based on our findings, we reflected on the direction of research of intelligent decision-making technology in wargaming environments.

In order to eliminate the influence of the high-order magnetic dipole modes in the short-wave range of the high-refractive-index dielectric grating, we propose a thin film filter to cut off the “useless” short-wave. In this way, the high-order magnetic dipole can be suppressed, and the quality of the structure color is highly improved after the thin film filter cut off the incident light in the non-resonant band. The combined application of the thin film filter and the grating filter not only avoids the problem of too big film layer thickness, but also reduces the manufacturing process cost. For example, the film system (basic film system 0.5 L H 0.5 L) and thickness of green, 15° and 45° incident film filters are 21 layers, 1.76 μm and 41 layers, 4.02 μm. The color coordinates corresponding to the calculation results occupy a large area on the Commission Internationale de l’Eclairage 1931 (CIE 1931) chromaticity diagram, which proves that this design scheme can effectively improve the structural color quality. This method obtains excellent theoretical simulation results. This has important implications for high-end imaging equipment and sensors.