Optimization and regularization of complex task decomposition for blind removal of multi-factor degradation

Most existing image restoration methods based on deep neural networks are developed for images which only degraded by a single degradation mode and imaging under an ideal condition. They cannot be directly used to restore the images degraded by multi-factor coupling. A complex task decomposition regularization optimization strategy (TDROS) is proposed to solve the problem. The restoration of images degraded by multi-factor coupling is a complex task that can be solved by separating these multiple factors, that is, breaking the complex task into numbers of simpler tasks to make the entire complex problem be overcome more easily. Motivated by this idea, the TDROS decomposes the complex task of image restoration into two sub-task: the potential task constrained by regularization and the main task for reconstructing high-definition images. In TDROS, the front of the neural network is focused on the restoration of images degraded by additive noise, while the other part of the network is focused mainly on the restoration of images degraded by blur. We applied the TDROS to an 11-layer convolutional neural network (CNN) and compared it with initial CNNs from the aspects of restoration accuracy and generalization ability. Based on these results, we used TDROS to design a novel network model for the restoration of atmospheric turbulence-degraded images. The experimental results demonstrate that the proposed TDROS can improve the generalization ability of the existing network more effectively than current popular methods, offering a better solution for the problem of severely degraded image restoration. Moreover, the TDROS concept provides a flexible framework for low-level visual complex tasks and can be easily incorporated into existing CNNs.     

自适应融合层级特征的混合退化图像复原算法

多种退化类型混合的图像比单一类型的退化图像降质更严重,很难建立精确模型对其复原,研究端到端的神经网络算法是复原的关键.现有的基于操作选择注意力网络的算法(operation-wiseattentionnetwork,OWAN)虽然有一定的性能提升,但是其网络过于复杂,运行较慢,复原图像缺乏高频细节,整体效果也有提升的空间.针对这些问题,提出一种基于层级特征融合的自适应复原算法.该算法直接融合不同感受野分支的特征,增强复原图像的结构;用注意力机制对不同层级的特征进行动态融合,增加模型的自适应性,降低了模型冗余;另外,结合L1损失和感知损失,增强了复原图像的视觉感知效果.在DIV2K,BSD500等数据集上的实验结果表明,该算法无论是在峰值信噪比和结构相似性上的定量分析,还是在主观视觉质量方面,均优于OWAN算法,充分证明了该算法的有效性.     

Incorporating microgrids coupling with utilization of flexible switching to enhance self-healing ability of electric distribution systems

Electric distribution networks have to deal with issues caused by natural disasters. These problems possess unique characteristics, and their severity can make load restoration methods impotent. One solution that can help in alleviating the aftermath is the use of microgrids (MGs). Employing the cumulative capacity of the generation resources through MG coupling facilitates the self-healing capability and leads to better-coordinated energy management during the restoration period, while the switching capability of the system should also be considered. In this paper, to form and schedule dynamic MGs in distribution systems, a novel model based on mixed-integer linear programming (MILP) is proposed. This approach employs graph-related theories to formulate the optimal formation of the networked MGs and management of their proper participation in the load recovery process. In addition, the Benders decomposition technique is applied to alleviate computability issues of the optimization problem. The validity and applicability of the proposed model are evaluated by several simulation studies.     

新时代长江鱼类多样性保护的思考

长江是我国第一大河流,是世界生物多样性的热点区域,也是我国鱼类多样性最高的河流之一。在介绍长江鱼类主要生态习性生境条件的基础上,分析了导致鱼类多样性下降的主要因素——生境丧失和过度捕捞,并简要总结了目前已采取的保护措施及存在的问题,提出了关于新时期长江鱼类多样性保护的思考与建议:①建立以流域管理机构牵头,相关部门和地方共同组建流域生态保护机制;②有计划开展支流及通江湖泊的生态恢复,使支流的小水电逐步退出,使原本的通江湖泊恢复自然江湖关系;③提高民众环保意识,积极建立全流域禁捕机制和休闲垂钓机制,并让垂钓者成为渔政的管理者和监督者。     

四川省植被覆盖时空演变及未来变化趋势分析

四川省地处长江上游,是长江流域重要的生态屏障;解析该区域植被覆盖的时空演变特征,对长江上游的生态环境修护与治理具有重要意义。基于长时间序列的MODIS NDVI数据,解析2000~2016年间四川省植被的空间分布格局与时间变化特征;采用趋势分析法和Hurst指数,预测未来植被的变化趋势及持续性。结果表明:(1)四川省NDVI多年平均值为0.50,植被覆盖整体上呈现川东高于川西的空间分布特征;(2)2000~2016年间,四川省的NDVI值在0.48至0.53之间波动,整体呈上升趋势,增长率为0.002 5/a;(3)植被覆盖度在未来变化将呈现改善大于退化的趋势;植被持续性改善的面积大于持续性退化的面积。     

Transfer learning between crop types for semantic segmentation of crops versus weeds in precision agriculture

Agricultural robots rely on semantic segmentation for distinguishing between crops and weeds to perform selective treatments and increase yield and crop health while reducing the amount of chemicals used. Deep‐learning approaches have recently achieved both excellent classification performance and real‐time execution. However, these techniques also rely on a large amount of training data, requiring a substantial labeling effort, both of which are scarce in precision agriculture. Additional design efforts are required to achieve commercially viable performance levels under varying environmental conditions and crop growth stages. In this paper, we explore the role of knowledge transfer between deep‐learning‐based classifiers for different crop types, with the goal of reducing the retraining time and labeling efforts required for a new crop. We examine the classification performance on three datasets with different crop types and containing a variety of weeds and compare the performance and retraining efforts required when using data labeled at pixel level with partially labeled data obtained through a less time‐consuming procedure of annotating the segmentation output. We show that transfer learning between different crop types is possible and reduces training times for up to 80%. Furthermore, we show that even when the data used for retraining are imperfectly annotated, the classification performance is within 2% of that of networks trained with laboriously annotated pixel‐precision data.     

Die Baugesellschaft Gebrüder Rank und ihre spanischen Ziegelgewölbe in München

The Rank construction company and the introduction of brick vaults in Munich after 1945 The construction of vaults without scaffolding with thin bricks has a centuries‐old tradition, particularly in Spain and Italy. Since the beginning of the 19^th century, it has been successfully spread mainly on the American double continent. The Rank Brothers construction company in Munich is known to date as one of those construction companies that began building with concrete in its home city very early on. New investigations, which were made possible above all by the provision of archive material by Paul Basiner, now point to an additional focus of this company's work. The architects and engineers of the company had been also able to familiarise themselves with this traditional Spanish brick‐vaulted construction method via the branch office in Spain, which was established from 1911 on. This knowledge was particularly in demand in the years following World War II, when the destroyed vaults had to be rebuilt with simple means and low material consumption. Together with Carl Sattler, who had become acquainted with similar construction methods in Italy, brick vaults were initially used to complete the construction of the Landeszentralbank in Munich. In the following years, Rank was able to provide many buildings in and around Munich with brick vaults. It was not until the 1960s that this construction technique went out of fashion and was thus somewhat forgotten.     

水下图像处理技术研究综述

随着自主式水下机器人的发展，水下探测技术成为新的研究热点。然而，吸收效应和散射效应导致水下获取的图像存在雾化和色彩偏差等缺陷。降质的水下图像在一定程度上降低了水下目标识别的准确性。为了改善水下图像质量，国内外学者对水下图像处理方法进行了深入研究。因水下图像处理方法对提升水下目标识别准确性具有良好的促进作用，故其具有重要的研究与分析价值。介绍了水下成像模型，分析了水下图像视觉质量下降的原理；根据水下物理成像模型将水下图像处理方法分为水下图像增强与水下图像复原，并分别对两类方法的研究现状进行分析与归纳；最后，总结与讨论了各类方法的优缺点，并展望了未来的发展方向。