Micro-expression recognition from local facial regions

MiE is a facial involuntary reaction that reflects the real emotion and thoughts of a human being. It is very difficult for a normal human to detect a Micro-Expression (MiE), since it is a very fast and local face reaction with low intensity. As a consequence, it is a challenging task for researchers to build an automatic system for MiE recognition. Previous works for MiE recognition have attempted to use the whole face, yet a facial MiE appears in a small region of the face, which makes the extraction of relevant features a hard task. In this paper, we propose a novel deep learning approach that leverages the locality aspect of MiEs by learning spatio-temporal features from local facial regions using a composite architecture of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM). The proposed solution succeeds to extract relevant local features for MiEs recognition. Experimental results on benchmark datasets demonstrate the highest recognition accuracy of our solution with respect to state-of-the-art methods.     

起·承·转·合 ——解读巍宝山长春洞空间布局的结构章法

长春洞是巍宝山保存最完整的宫 观，虽占地不到两亩①，却在有限的空间中 创造出九楼十院的空间格局。本文以长春洞 空间布局序列为研究对象，从知觉体验的 连续性入手，提出了空间变化对参观者情感 的影响，论述了在建筑空间布局序列中参观 者知觉体验逐层递进的控制机制，探讨了以 “起、承、转、合”为结构，以“知觉体验”为 章法的动态空间布局设计手法，希望为道教 建筑空间与其他超越日常的精神空间的研 究提供启示。     

User activities and the heterogeneity of urban space:The case of Dahiyat Al Hussein park

Urban researchers have maintained a constant interest in the complexity and continuity of urban space usage. Some have applied actor–network theory (ANT) to investigate the heterogeneity of spaces and present them through the networks of their users’ activities. However, these accounts are predominantly limited in examining the extent to which these spaces may be heterogeneous when exploring such networks. This paper draws on recent ANT scholarship, which employs an ethnographic research conducted in a main park in a housing project at Dahiyat Al Hussein in Amman, Jordan. The findings describe the complex and unpredictable negotiations that occur within spaces by documenting the varieties and interrelations among user activity networks within this common and shared urban space. This research reveals the extent to which spaces, parks in this case, may be heterogeneous by unpacking their usage. The conclusions and insights assert the necessity of paying attention to design detail and creating designs that are responsive to evolving user activities.     

On the equilibrium of funicular polyhedral frames and convex polyhedral force diagrams

This paper presents a three-dimensional extension of graphic statics using polyhedral form and force diagrams for the design of compression-only and tension-only spatial structures with externally applied loads. It explains the concept of 3D structural reciprocity based on Rankine’s original proposition for the equilibrium of spatial frames. It provides a definition for polyhedral reciprocal form and force diagrams that allows including external forces and discusses their geometrical and topological characteristics. This paper furthermore provides a geometrical procedure for constructing a pair of reciprocal polyhedral diagrams from a given polyhedron representing either the form or force diagram of a structural system. Using this method, this paper furthermore suggests a design strategy for finding complex funicular spatial forms in pure compression (or tension), based on the construction of force diagrams through the aggregation of convex polyhedral cells. Finally, it discusses the effect of changes in the geometry of the force diagram on the geometry of the form diagram and the distribution of forces in it.     

Spatial filtering velocimeter using frequency shifting by the method of rotating kernel

A new approach of frequency shifting by rotating kernel is proposed to improve the performance of a spatial filtering velocimeter, used to provide accurate velocity information for a vehicle self-contained navigation system. A linear CMOS image sensor was employed both as a spatiotemporal differential spatial filter and as a photodetector. The filtering operation was fully performed in FPGA and is realized by applying a rotating kernel to the pixel values of the image. Theoretical analysis showed this method could double the maximum measurable velocity. The power spectrum of the output signal was obtained by fast Fourier transform (FFT), and was corrected by a frequency spectrum correction algorithm, named energy centrobaric correction. This velocimeter was used to measure the moving velocities of a conveyor belt. Experimental results verified the method’s ability of reducing the output signal frequency and standard uncertainty of velocity measurement. What is more, the undesired output introduced by frequency shifting to the power spectrum of the output signal was deeply investigated and a new method was proposed to eliminate the undesired component in output signals. This velocimeter aims at providing accurate velocity information for vehicle autonomous navigation system.     

A geospatial framework to support integrated biogeochemical modelling in the United Kingdom

Anthropogenic impacts on the aquatic environment, especially in the context of nutrients, provide a major challenge for water resource management. The heterogeneous nature of policy relevant management units (e.g. catchments), in terms of environmental controls on nutrient source and transport, leads to the need for holistic management. However, current strategies are limited by current understanding and knowledge that is transferable between spatial scales and landscape typologies. This study presents a spatially-explicit framework to support the modelling of nutrients from land to water, encompassing environmental and spatial complexities. The framework recognises nine homogeneous landscape units, distinct in terms of sensitivity of nutrient losses to waterbodies. The functionality of the framework is demonstrated by supporting an exemplar nutrient model, applied within the Environmental Virtual Observatory pilot (EVOp) cloud cyber-infrastructure. We demonstrate scope for the use of the framework as a management decision support tool and for further development of integrated biogeochemical modelling.     

Seasonal movements of muskellunge in the St. Clair – Detroit River System: Implications for multi-jurisdictional fisheries management

The St. Clair-Detroit River System contains a world-class Great Lakes muskellunge (Esox masquinongy) fishery that has avoided the declines observed in many Great Lakes muskellunge populations. Muskellunge are an upper trophic level predator, and therefore a naturally low-density species. Limited fishery-independent data exist on which to base management decisions. To remedy this, we initiated an acoustic telemetry study in May of 2016, in collaboration with the Great Lakes Acoustic Telemetry Observation System. Our objective was to describe patterns of movement of muskellunge in this large and open system to better understand their spatial ecology. We acoustically tagged 133 muskellunge in the Detroit River and Lake St. Clair, and movements of 58 fish that passed our data quality control screens were analyzed. We utilized mixed modelling to assess the effects of sex, length, release location, and season on daily movement rates. We found that movement rates only differed among seasons, with highest movement rates occurring in the fall and lowest movement rates in the winter. Muskellunge tagged at different locations exhibited distinct residency patterns, and fish frequently crossed jurisdictional and waterbody boundaries. Ultimately our study highlights the scope and patterns of muskellunge movement in a large, unimpounded system and demonstrates that management of these fish would benefit from consideration of their full distribution covering multiple management jurisdictions.     

A distance-type-insensitive clustering approach

In this paper, we offer a method aiming to minimize the role of distance metric used in clustering. It is well known that distance metrics used in clustering algorithms heavily influence the end results and also make the algorithms sensitive to imbalanced attribute/feature scales. To solve these problems, a new clustering algorithm using a per-attribute/feature ranking operating mechanism is proposed in this paper. Ranking is a rarely used discrete, nonlinear operator by other clustering algorithms. However, it also has unique advantages over the dominantly used continuous operators. The proposed algorithm is based on the ranks of the data samples in terms of their spatial separation and is able to provide a more objective clustering result compared with the alternative approaches. Numerical examples on benchmark datasets prove the validity and effectiveness of the proposed concept and principles.     

伽马刀准直体三维立体定向孔的加工方法

伽马刀准直体上三维立体定向孔的聚焦精度是影响治疗效果的关键因素,也是加工的难点。介绍了伽马刀准直体三维立体定向孔的加工方法,设计了工装,并研究了关键精度控制的问题。