Underwater image super-resolution using multi-stage information distillation networks

Recently, single image super-resolution (SISR) has been widely applied in the fields of underwater robot vision and obtained remarkable performance. However, most current methods generally suffered from the problem of a heavy burden on computational resources with large model sizes, which limited their real-world underwater robotic applications. In this paper, we introduce and tackle the super resolution (SR) problem for underwater robot vision and provide an efficient solution for near real-time applications. We present a novel lightweight multi-stage information distillation network, named MSIDN, for better balancing performance against applicability, which aggregates the local distilled features from different stages for more powerful feature representation. Moreover, a novel recursive residual feature distillation (RRFD) module is constructed to progressively extract useful features with a modest number of parameters in each stage. We also propose a channel interaction & distillation (CI&D) module that employs channel split operation on the preceding features to produce two-part features and utilizes the inter channel-wise interaction information between them to generate the distilled features, which can effectively extract the useful information of current stage without extra parameters. Besides, we present USR-2K dataset, a collection of over 1.6K samples for large-scale underwater image SR training, and a testset with an additional 400 samples for benchmark evaluation. Extensive experiments on several standard benchmark datasets show that the proposed MSIDN can provide state-of-the-art or even better performance in both quantitative and qualitative measurements.     

基于光刀图像的机器人示教纠偏方法

为了解决机器人辅助在线检测系统示教过程中效率低、人为干预多的问题,针对孔类特征,提出一种通过处理光刀图像获取机器人位姿纠正参数的方法。在该方法中,纠偏过程按照将图像光刀线调整为水平方向、将被测特征调整至图像中央、根据不同特征调整最优扫描方向3个步骤进行。根据光刀图像可以求出测量坐标系与被测特征的位姿关系,并得到该位姿与理想测量位姿的偏差。经过坐标变换即可得到在机器人基坐标系下的调整参数。另外,可利用电脑控制机器人运动,从而实现了纠偏过程的自动化。实验结果表明,该方法可将初始的示教位姿修正为理想位姿,且比传统方式效率更高。     

Sub-band coding of hexagonal images

According to the circle-packing theorem, the packing efficiency of a hexagonal lattice is higher than an equivalent square tessellation. Consequently, in several contexts, hexagonally sampled images compared to their Cartesian counterparts are better at preserving information content. In this paper, novel mapping techniques alongside the wavelet compression scheme are presented for hexagonal images. Specifically, we introduce two tree-based coding schemes, referred to as SBHex (spirally-mapped branch-coding for hexagonal images) and BBHex (breadth-first block-coding for hexagonal images). Both of these coding schemes respect the geometry of the hexagonal lattice and yield better compression results. Our empirical results show that the proposed algorithms for hexagonal images produce better reconstruction quality at low bits per pixel representations compared to the tree-based coding counterparts for the Cartesian grid.     

Hydrogen production via ammonia decomposition catalyzed by Ni/M–Mo–N (M = Ni,Co) bimetallic nitrides

This study demonstrates the significant improvement in NH₃ decomposition using Ni-decorated M–Mo–N-based catalysts (M = Co and Ni) compared with conventional catalysts. Catalysts are prepared using a mixture of the corresponding metal salts and hexamethylenetetramine, and the impregnation method is used to decorate the Ni-particles on the catalysts. Among all the samples, 10 wt% Ni-decorated Co₃Mo₃N exhibits the highest NH₃ conversion rate (71%) at 500 °C, and the performance remains stable for 30 h of long-term testing. According to the gas chromatography measurements, the H₂/N₂ ratio is approximately 3 in all cases, which is consistent with the theoretical value. X-ray photoelectron spectroscopy results show that Co₃Mo₃N possesses the highest NH₃ conversion efficiency because of the weaker binding energy of Mo–N. Furthermore, Co₃Mo₃N exhibits a stronger Lewis acidity and higher NH₃ decomposition, which is attributed to the easy breaking of the N–H bond on the Co₃Mo₃N surface.     

One-step synthesis of Ni/yttrium-doped barium zirconates catalyst for on-site hydrogen production from NH3 decomposition

On-site produced hydrogen from ammonia decomposition can directly fuel solid oxide fuel cells (SOFCs) for power generation. The key issue in ammonia decomposition is to improve the activity and stability of the reaction at low temperatures. In this study, proton-conducting oxides, Ba(Zr,Y) O_3-δ (BZY), were investigated as potential support materials to load Ni metal by a one-step impregnation method. The influence of Ni loading, Ba loading, and synthesis temperature, of Ni/BZY catalysts on the catalytic activity for ammonia decomposition were investigated. The Ni/BZY catalyst with Ba loading of 20 wt%, Ni loading of 30 wt%, and synthesized at 900 °C attained the highest ammonia conversion of 100% at 600 °C. The kinetics analysis revealed that for Ni/BZY catalyst, the hydrogen poisoning effect for ammonia decomposition was significantly suppressed. The reaction order of hydrogen for the optimized Ni/BZY catalyst was estimated as low as ?0.07, which is the lowest to the best of our knowledge, resulting in the improvement in the activity. H₂ temperature programmed reduction and desorption analysis results suggested that a strong interaction between Ni and BZY support as well as the hydrogen storage capability of the proton-conducting support might be responsible for the promotion of ammonia decomposition on Ni/BZY. Based on the experimental data, a mechanism of hydrogen spillover from Ni to BZY support is proposed.     

Deterministic and probabilistic ship pitch prediction using a multi-predictor integration model based on hybrid data preprocessing,reinforcement learning and improved QRNN

The deterministic and probabilistic prediction of ship motion is important for safe navigation and stable real-time operational control of ships at sea. However, the volatility and randomness of ship motion, the non-adaptive nature of single predictors and the poor coverage of quantile regression pose serious challenges to uncertainty prediction, making research in this field limited. In this paper, a multi-predictor integration model based on hybrid data preprocessing, reinforcement learning and improved quantile regression neural network (QRNN) is proposed to explore the deterministic and probabilistic prediction of ship pitch motion. To validate the performance of the proposed multi-predictor integrated prediction model, an experimental study is conducted with three sets of actual ship longitudinal motions during sea trials in the South China Sea. The experimental results indicate that the root mean square errors (RMSEs) of the proposed model of deterministic prediction are 0.0254°, 0.0359°, and 0.0188°, respectively. Taking series #2 as an example, the prediction interval coverage probabilities (PICPs) of the proposed model of probability predictions at 90%, 95%, and 99% confidence levels (CLs) are 0.9400, 0.9800, and 1.0000, respectively. This study signifies that the proposed model can provide trusted deterministic predictions and can effectively quantify the uncertainty of ship pitch motion, which has the potential to provide practical support for ship early warning systems.     

基于改进SOA-VMD的磁声发射信号去噪算法

磁声发射（MAE）是铁磁性材料磁化过程中产生的声发射信号，在构件应力检测和微观损伤检测中有着广泛的应用。针对MAE信号非稳态、复杂性、衰减性等特点，提出海鸥算法结合变分模态分解（SOA-VMD）的去噪方法，为克服海鸥算法求解过程中易陷入局部最优解问题，利用柯西变异算子产生随机迭代过程，使改进算法即柯西变异海欧算法（CVSOA）跳出早熟收敛。采用以幅值谱熵为适应度函数，优化VMD算法中分解模态个数K和二次惩戒因子α两个参数，将含噪声的MAE信号进行VMD分解重构。经仿真信号和实际检测信号分析表明，改进后的CVSOA-VMD算法全局寻优能力和去噪性能优于传统的SOA-VMD算法，降噪后的MAE信号特征值对于不同应力下均方根、偏斜度特征值的重复性更好，可靠性更高。     

基于改进残差特征蒸馏的轻量级超分辨率网络

基于深度学习的图像超分辨率算法通常采用递归的方式或参数共享的策略来减少网络参数，这将增加网络的深度，使得运行网络花费大量的时间，从而很难将模型部署到现实生活中。为了解决上述问题，本文设计一种轻量级超分辨率网络，对中间特征的关联性及重要性进行学习，且在重建部分结合高分辨率图像的特征信息。首先，引入层间注意力模块，通过考虑层与层之间的相关性，自适应地分配重要层次特征的权重。其次，使用增强重建模块提取高分辨率图像中更精细的特征信息，以此得到更加清晰的重建图片。通过大量的对比实验表明，本文设计的网络与其他轻量级模型相比，有更小的网络参数量，并且在重建精度和视觉效果上都有一定的提升。     

Analytical approach on magnetohydrodynamic Casson fluid flow past a stretching sheet via Adomian decomposition method

Flow phenomena of three-dimensional conducting Casson fluid through a stretching sheet are proposed in the present investigation with the impact of the magnetic parameter in a permeable medium. The adaptation of particular transformations is useful to modify the governing equations into their nondimensional as well as the ordinary form. However, these transformed equations are nonlinear and approximate analytical methods for the solution of the complex form of governing equations. In particular, the Adomian decomposition method is proposed for the solution. The behavior of several variables, such as the magnetic and porous matrix, on the flow profile as well as the rate of shear stress, are discussed via graphs and tables. The conformity of the current result with the earlier study shows a road map for further investigation. The major concluding remarks are; the retardation in the velocity distribution is rendered due to an increase in the Casson parameter moreover, the Casson parameter favors in reducing the rate of shear stress coefficient in magnitude.