The goal of Advance Traveler Information System (ATIS) is to improve traffic flow and safety by providing up-to-date information of traffic network. In this paper, traffic information (travel time) estimation based on fused traffic state data is presented. A centralized architecture is used to fuse the traffic sate data from different sensors based fusion by averaging and fusion by median and to estimate the travel time based on the simultaneous travel time estimation model accordingly. Two case studies are selected to investigate the performance of fusion models based on freeway data in USA and the arterial road data in South Korea. The results show that the fusion by median performs best. The model is able to eliminate outliers in the data with less effort of complex mathematical process. It can be used as a benchmark for comparison with other advanced fusion models.

抄録全体を表示

Marcel Duchamp (1887-1968) was in contact with the Salon Cubists in his cubist period (1911-1912). Unlike Picasso and Braque, they formulated some theories based on the new concepts of philosophy, literature, and physics in the early 20th century: Time, Simultaneity, and the Fourth Dimension. These concepts are related to their Platonistic quest for , and and Simultaneity functioned as the ways to represent the invisible truth, the Fourth Dimension. Thus it is conceivable that Duchamp also sympathized with their quest for truth. In the Cubists' paintings the fragmentation of an object caused by Simultaneity results in the destruction of a narrative sequence in the traditional paintings; on the contrary, Duchamp's paintings like Nude Descending a Staircase, No.2 carry it because he depicts an autonomous movement of an object, and a painter sees it from a static viewpoint. Furthermore, he depends on the relationship between words and painting to imply the Fourth Dimension. Consequently, I concluded that the uniqueness of Duchamp is that he never excluded the external elements such as words from painting; In other words, he still carried the viewpoint of a traditional painter.

抄録全体を表示

In recent years, driver's visual attention has been actively studied for driving automation technology. However, the number of models is few to perceive an insight understanding of driver's attention in various moments. All attention models process multi-level image representations by a two-stream/multi-stream network, increasing the computational cost due to an increment of model parameters. However, multi-level image representation such as optical flow plays a vital role in tasks involving videos. Therefore, to reduce the computational cost of a two-stream network and use multi-level image representation, this work proposes a single stream driver's visual attention model for a critical situation. The experiment was conducted using a publicly available critical driving dataset named BDD-A. Qualitative results confirm the effectiveness of the proposed model. Moreover, quantitative results highlight that the proposed model outperforms state-of-the-art visual attention models according to CC and SIM. Extensive ablation studies verify the presence of optical flow in the model, the position of optical flow in the spatial network, the convolution layers to process optical flow, and the computational cost compared to a two-stream model.

抄録全体を表示

For hyperspectral imaging, we proposed an edge-preserving filter, named hyperspectral Gaussian filtering, and its separable implementation for accelerating the proposed filter. Experimental results show that the acceleration has superior performance to the brute-force implementation and the other state-of-the-art methods in denosing. Also, we successfully achieves 70-times speedup with the acceleration.

抄録全体を表示

In the aquaculture industry, damage occurs because of a sudden decrease in salinity concentration. Therefore, the demand for real-time forecasting has increased. Forecasting through machine learning is increasing; however, observation stations at the target site are not always present. Therefore, we predicted the flow field at the target site through data assimilation (DA) using a method combining the Kalman filter and finite element method. In this study, we used the predicted values with DA for long short-term memory and improved the prediction accuracy.

抄録全体を表示

Subspace representation model is an important subset of visual tracking algorithms. Compared with models performed on the original data space, subspace representation model can effectively reduce the computational complexity, and filter out high dimensional noises. However, for some complicated situations, e.g., dramatic illumination changing, large area of occlusion and abrupt object drifting, traditional subspace representation models may fail to handle the visual tracking task. In this paper, we propose a novel subspace representation algorithm for robust visual tracking by using low-rank representation with graph constraints (LRGC). Low-rank representation has been well-known for its superiority of handling corrupted samples, and graph constraint is flexible to characterize sample relationship. In this paper, we aim to exploit benefits from both low-rank representation and graph constraint, and deploy it to handle challenging visual tracking problems. Specifically, we first propose a novel graph structure to characterize the relationship of target object in different observation states. Then we learn a subspace by jointly optimizing low-rank representation and graph embedding in a unified framework. Finally, the learned subspace is embedded into a Bayesian inference framework by using the dynamical model and the observation model. Experiments on several video benchmarks demonstrate that our algorithm performs better than traditional ones, especially in dynamically changing and drifting situations.

抄録全体を表示

In this study a new type of ensemble forecast assimilation technique is developed in order to improve the forecast skill in the nonlinear dynamical system. The forecast assimilation is an analysis technique in which a true value contained in each ensemble forecast is accumulated into a single assimilated forecast such as a data assimilation. For the experiments, we used a Lorenz model, and a Kalman filter is applied for the forecast assimilation.
The experiments are started by calculating 101 members of the ensemble forecast in which the initial error with Gaussian distribution is superimposed around the truth, and one of the members is arbitrarily selected as a control forecast. The experiments of the forecast assimilation are repeated 5000 times for different sectors of the solution trajectory to obtain the statistical significance of the results. The distribution of the ensemble members is stretched by a linear error growth at the beginning of the forecast. After that, the nonlinear effect becomes dominant to distort the distribution. The forecast assimilation is then started when the errors of the ensemble forecasts have grown to a certain threshold. It is demonstrated that the forecast skill of the assimilated forecast is always superior to the control forecast. In the range of the small root mean square error (RMSE) of the ensemble forecast, the skill of the assimilated forecast is inferior to the ordinary ensemble mean. However, for the sufficiently large RMSE before the saturation, it is shown that the skill of the assimilated forecast is superior to the ensemble mean. The result suggests that the forecast assimilation is one of the viable approaches to the medium or extended range forecast.

抄録全体を表示

Learning collocations is a crucial concern in the development of English communications skills. However, learners often feel that collocations are difficult to memorize. Hence, providing more user-friendly learning material for students is necessary. The possibility of teaching basic collocations for Japanese students through movies is investigated using 1,572 basic collocations and a database of movie quotes for 77 famous movies. Three hundred and sixty-two collocations have been included in the database. The movies containing the maximum numbers of collocations are Erin Brockovich, Gosford Park, Anne of Green Gables, Jackie Brown, and Super Size Me. These movies have maximum 31 and minimum 26 kinds of basic collocations, respectively. The collocations appearing in several movies, having been observed in 22 to 54 works, are take care, do thing, take look, do favor, and tell truth. This indicates that using movies does not ensure that there will be a sufficient number of collocations to be presented. Consequently, merely through using movies, teachers cannot present sufficient volume of collocations to be learned. Thus, movies must be used in combination of other teaching materials and learning methods such as word lists and textbooks.

抄録全体を表示

This paper presents a real-time beat-tracking method that integrates audio and visual information in a probabilistic manner to enable a humanoid robot to dance in synchronization with music and human dancers. Most conventional music robots have focused on either music audio signals or movements of human dancers to detect and predict beat times in real time. Since a robot needs to record music audio signals with its own microphones, however, the signals are severely contaminated with loud environmental noise. To solve this problem, we propose a state-space model that encodes a pair of a tempo and a beat time in a state-space and represents how acoustic and visual features are generated from a given state. The acoustic features consist of tempo likelihoods and onset likelihoods obtained from music audio signals and the visual features are tempo likelihoods obtained from dance movements. The current tempo and the next beat time are estimated in an online manner from a history of observed features by using a particle filter. Experimental results show that the proposed multi-modal method using a depth sensor (Kinect) to extract skeleton features outperformed conventional mono-modal methods in terms of beat-tracking accuracy in a noisy and reverberant environment.

抄録全体を表示

In response to the poor performance of Visual SLAM in low light scenes and its high requirements for real-time computing, we propose a 2D-3D fused neural network that processes time consistency with minimal computational cost for enhancing a sequence of low-light images, and design two new loss functions to guide the network to focus on enhancing corners and edges. We then integrated our method into VINS-Mono and compared it with several low-light enhancement methods. The results show that our method achieves approximately 30% higher positioning accuracy than previous researches, while also having the shortest inference time.

抄録全体を表示

Computation tree logic (CTL) is known to be one of the most useful temporal logics for verifying concurrent systems by model checking technologies. However, CTL is not sufficient for handling inconsistency-tolerant and probabilistic accounts of concurrent systems. In this paper, a paraconsistent (or inconsistency-tolerant) probabilistic computation tree logic (PpCTL) is derived from an existing probabilistic computation tree logic (pCTL) by adding a paraconsistent negation connective. A theorem for embedding PpCTL into pCTL is proven, thereby indicating that we can reuse existing pCTL-based model checking algorithms. A relative decidability theorem for PpCTL, wherein the decidability of pCTL implies that of PpCTL, is proven using this embedding theorem. Some illustrative examples involving the use of PpCTL are also presented.

抄録全体を表示

In the field of biology, behavior analysis a valuable tool for observing the change and development of the brain and neural mechanism non-invasively. However, behavior analysis has enormous data processing because of lack in automatic tools for the measurement of behavioral data. Therefore, the purpose of this study is to develop the new motion capture system which enables us to improve processing a vast amount of tracking data about chicks behavior. Specifically, we propose automatic motion capture system obtaining the position and direction of chicks by using video movie from only WEB camera before and after fluttering of chicks.

抄録全体を表示

Paratransit service has characteristics such as no stopping spot, no timetable, and irregular pattern because stopping depends on driver or passenger demand. This study uses Wi-Fi scanner equipment to capture media access control (MAC) address of the paratransit passenger in Makassar city, Indonesia. The objective of this study is 1) to produce a cleaning procedure to clean Wi-Fi raw data from non-passenger data, 2) to match data between ground truth and Wi-Fi, and 3) to make an OD or boarding alighting data based on Wi-Fi estimated. The result showed that 1211 of the 2925 real passengers from Wi-Fi data after MAC address processing matched with ground truth data, and the possible similarity in the number of actual passenger and MAC address is 41.4%. The trend between Wi-Fi cleaning result and ground truth is similar and the same tendency. The correlation value between Wi-Fi cleaning result and ground truth is 0.89.

抄録全体を表示

A content clause is a that-clause that is in apposition to a noun and expresses its content. The present paper is an attempt to consider what kinds of nouns can take a content clause and to explain how the content clause construction is related to some other similar constructions. It will be claimed that nouns interpretable as something composed of words can take a content clause and that the CONTAINER-CONTENT image-schema suggests a perspective from which to relate the content clause construction to the relative (adverbial) clause construction.

抄録全体を表示

The objective of this paper is to examine travel time variability and reliability on an urban arterial road. Towards this purpose, travel time data was collected using Wi-Fi sensors for fourweeks. Generalized extreme value (GEV) distribution was observed as the best-fitted distribution for explaining variations in travel time values. The variations in travel time werealso assessed by the time of the day (TOD) and day of the week (DOW). The results obtained indicate that travel time variation is significantly influenced by TOD and DOW. It was observed that buffer time index (BTI) explained variations in travel time better compared to other reliability parameters such as planning time index (PTI) and travel time index (TTI). As an important outcome, the study developed reliability-based level of service thresholds (LOS), which can enable traffic engineers to evaluate the performance of an urban arterial road based on travel time variability and reliability.

抄録全体を表示

マルチエージェントシミュレーションによって群集の移動を計算する研究が行われてい る.このようなシミュレーションでは各エージェントに経路や帰宅開始時刻などのシナリオを与える ことでその後の群集の移動を計算し帰宅人数の推移などを推定することが可能である.しかしなが ら，様々なシナリオを網羅的にシミュレーションすることは，計算資源の観点から限界がある.この 問題に対して，本研究では各エージェントのシナリオと初期座標，途中までのシミュレーション結果 から，その後の結果を推定するニューラルネットワークを提案する.実験の結果，提案手法は短時間 かつ高精度でシミュレーション結果を推定できることが分かった.この知見は，提案手法によってシ ミュレーションの計算過程の一部を代替し，シミュレーションの実行時間を短縮できることを示して いる.

抄録全体を表示

This study examines the impacts of political direct mail on grassroots activism by investigating the fundraising drives of the Barry Goldwater campaign in the 1964 presidential election. Goldwater was the first conservative politician nominated by the Republican Party. With enthusiastic support from the right wing, especially anticommunist organizations such as the John Birch Society (JBS), the Goldwater campaign indicated that anti-liberalism could be a national political force in the United States. Throughout the 1964 election, Goldwaterites contacted millions of voters, built up the collective identity of conservatives, and suggested the GOP would be able to make gains in the West and the South. Despite the resounding defeat on Election Day, the Goldwater movement set down the groundwork for the rise of conservatism in the following decades.

Goldwater’s fundraising campaign also transformed a long-term pattern in political contribution. Direct mail was a new marketing technique of the advertising industry in the postwar era. Unlike mass media including newspaper, radio, and television, direct mail functioned as a personalized medium which distributed different message to individuals, instead of standardized information to the masses. By sending out millions of fundraising letters, Goldwater carved out political niches and constructed a huge database of conservative Americans during the election. Simultaneously, direct mail solicitation changed the amount of each contribution. Prior to the 1960s, both major parties had depended largely on big contributions by business interests and philanthropists. Yet direct mail made it possible to tap small funds such as one or five dollars, opening the door for small money politics. After the Goldwater campaign successfully received the mass of small contributions, Democrats and Republicans began to raise funds from ordinary voters.

This article explores the 1964 Goldwater movement, focusing on two right-wing groups. The JBS mobilized many middle-class Americans in suburban areas throughout the Sunbelt, and a cadre of New York conservatives, including conservative intellectual William F. Buckley Jr., National Review publisher William A. Rusher, and political consultant Marvin Liebman, carried out advertising campaigns on Goldwater’s behalf. Whereas the JBS encouraged the grassroots to join the movement through local chapters across the nation, the New York conservatives reached out to individuals with direct mailings. Contrasting the two types of activism, this study analyzes the interactions between the leadership and the grassroots, as well as controversies within the conservative movement in 1964.

The 1964 presidential election demonstrates that Goldwater’s direct mail added a new definition of grassroots activism in political campaigns. As Alexis de Tocqueville observed in the 1830s, associations and communities characterized American democracy. However, direct mail as the personalized medium directly connected the leadership to the grassroots without any censorship and regulation, and the information technology recast the grassroots from “activism based on face-to-face relationship” toward “accumulation of small involvements.” Thus, the Goldwater campaign not only galvanized the conservative movement but also influenced political participation in 1960s America.

抄録全体を表示