A super-resolution algorithm is proposed to solve the problem of measuring the thin thickness of oil slick using compressed sensing theory. First, a mathematical model of a single pulse underwater ultrasonic echo is established. Then, the estimation model of the transmit time of flight (TOF) of ultrasonic echo within oil slick is given based on the sparsity of echo signals. At last, the super-resolution TOF value can be obtained by solving the sparse convex optimization problem. Simulations and experiments are conducted to validate the performance of the proposed method.

High-Altitude Platform Station (HAPS) provides communication services from an altitude of 20km via a stratospheric platform such as a balloon, solar-powered airship, or other aircraft, and is attracting much attention as a new mobile communication platform for ultra-wide coverage areas and disaster-resilient networks. HAPS can provide mobile communication services directly to the existing smartphones commonly used in terrestrial mobile communication networks such as Fourth Generation Long Term Evolution (4G LTE), and in the near future, Fifth Generation New Radio (5G NR). In order to design efficient HAPS-based cell configurations, we need a radio wave propagation model that takes into consideration factors such as terrain, vegetation, urban areas, suburban areas, and building entry loss. In this paper, we propose a new vegetation loss model for Recommendation ITU-R P.833-9 that can take transmission frequency and seasonal characteristics into consideration. It is based on measurements and analyses of the vegetation loss of deciduous trees in different seasons in Japan. Also, we carried out actual stratospheric measurements in the 700MHz band in Kenya to extend the lower frequency limit. Because the measured results show good agreement with the results predicted by the new vegetation loss model, the model is sufficiently valid in various areas including actual HAPS usage.

Traditional pipeline methods for task-oriented dialogue systems are designed individually and expensively. Existing memory augmented end-to-end methods directly map the inputs to outputs and achieve promising results. However, the most existing end-to-end solutions store the dialogue history and knowledge base (KB) information in the same memory and represent KB information in the form of KB triples, making the memory reader's reasoning on the memory more difficult, which makes the system difficult to retrieve the correct information from the memory to generate a response. Some methods introduce many manual annotations to strengthen reasoning. To reduce the use of manual annotations, while strengthening reasoning, we propose a hierarchical memory model (HM2Seq) for task-oriented systems. HM2Seq uses a hierarchical memory to separate the dialogue history and KB information into two memories and stores KB in KB rows, then we use memory rows pointer combined with an entity decoder to perform hierarchical reasoning over memory. The experimental results on two publicly available task-oriented dialogue datasets confirm our hypothesis and show the outstanding performance of our HM2Seq by outperforming the baselines.

Machine reading comprehension with multi-hop reasoning always suffers from reasoning path breaking due to the lack of world knowledge, which always results in wrong answer detection. In this paper, we analyze what knowledge the previous work lacks, e.g., dependency relations and commonsense. Based on our analysis, we propose a Multi-dimensional Knowledge enhanced Graph Network, named MKGN, which exploits specific knowledge to repair the knowledge gap in reasoning process. Specifically, our approach incorporates not only entities and dependency relations through various graph neural networks, but also commonsense knowledge by a bidirectional attention mechanism, which aims to enhance representations of both question and contexts. Besides, to make the most of multi-dimensional knowledge, we investigate two kinds of fusion architectures, i.e., in the sequential and parallel manner. Experimental results on HotpotQA dataset demonstrate the effectiveness of our approach and verify that using multi-dimensional knowledge, especially dependency relations and commonsense, can indeed improve the reasoning process and contribute to correct answer detection.

Autonomous vehicles and advanced driver assistant systems (ADAS) are receiving notable attention as research fields in both academia and private industry. Some decision-making systems use sets of logical rules to map knowledge of the ego-vehicle and its environment into actions the ego-vehicle should take. However, such rulesets can be difficult to create — for example by manually writing them — due to the complexity of traffic as an operating environment. Furthermore, the building blocks of the rules must be defined. One common solution to this is using an ontology specifically aimed at describing traffic concepts and their hierarchy. These ontologies must have a certain expressive power to enable construction of useful rules. We propose a process of generating sets of explanatory rules for ADAS applications from data using ontology as a base vocabulary and present a ruleset generated as a result of our experiments that is correct for the scope of the experiment.

When applying estimation methods, the issue of outliers is inevitable. The extent of their influence has not been clarified, though several studies have evaluated outlier elimination methods. It is unclear whether we should always be sensitive to outliers, whether outliers should always be removed before estimation, and what amount of precaution is required for collecting project data. Therefore, the goal of this study is to illustrate a guideline that suggests how sensitively we should handle outliers. In the analysis, we experimentally add outliers to three datasets, to analyze their influence. We modified the percentage of outliers, their extent (e.g., we varied the actual effort from 100 to 200 person-hours when the extent was 100%), the variables including outliers (e.g., adding outliers to function points or effort), and the locations of outliers in a dataset. Next, the effort was estimated using these datasets. We used multiple linear regression analysis and analogy based estimation to estimate the development effort. The experimental results indicate that the influence of outliers on the estimation accuracy is non-trivial when the extent or percentage of outliers is considerable (i.e., 100% and 20%, respectively). In contrast, their influence is negligible when the extent and percentage are small (i.e., 50% and 10%, respectively). Moreover, in some cases, the linear regression analysis was less affected by outliers than analogy based estimation.

Advances in deep reinforcement learning have demonstrated its effectiveness in a wide variety of domains. Deep neural networks are capable of approximating value functions and policies in complex environments. However, deep neural networks inherit a number of drawbacks. Lack of interpretability limits their usability in many safety-critical real-world scenarios. Moreover, they rely on huge amounts of data to learn efficiently. This may be suitable in simulated tasks, but restricts their use to many real-world applications. Finally, their generalization capability is low, the ability to determine that a situation is similar to one encountered previously. We present a method to combine external knowledge and interpretable reinforcement learning. We derive a rule-based variant version of the Sarsa(λ) algorithm, which we call Sarsa-rb(λ), that augments data with prior knowledge and exploits similarities among states. We demonstrate that our approach leverages small amounts of prior knowledge to significantly accelerate the learning in multiple domains such as trading or visual navigation. The resulting agent provides substantial gains in training speed and performance over deep q-learning (DQN), deep deterministic policy gradients (DDPG), and improves stability over proximal policy optimization (PPO).

Some non-acoustic modalities have the ability to reveal certain speech attributes that can be used for synthesizing speech signals without acoustic signals. This study validated the use of ultrasonic Doppler frequency shifts caused by facial movements to implement a silent speech interface system. A 40kHz ultrasonic beam is incident to a speaker's mouth region. The features derived from the demodulated received signals were used to estimate the speech parameters. A nonlinear regression approach was employed in this estimation where the relationship between ultrasonic features and corresponding speech is represented by deep neural networks (DNN). In this study, we investigated the discrepancies between the ultrasonic signals of audible and silent speech to validate the possibility for totally silent communication. Since reference speech signals are not available in silently mouthed ultrasonic signals, a nearest-neighbor search and alignment method was proposed, wherein alignment was achieved by determining the optimal pair of ultrasonic and audible features in the sense of a minimum mean square error criterion. The experimental results showed that the performance of the ultrasonic Doppler-based method was superior to that of EMG-based speech estimation, and was comparable to an image-based method.

Since software becomes more complex during its life cycle, the verification cost becomes higher, especially for such methods which are using model checking in general and assume-guarantee reasoning in specific. To address the problem of reducing the assume-guarantee verification cost, this paper presents a method to generate locally minimum and strongest assumptions for verification of component-based software. For this purpose, we integrate a variant of membership queries answering technique to an algorithm which considers candidate assumptions that are smaller and stronger first, larger and weaker later. Because the algorithm stops as soon as it reaches a conclusive result, the generated assumptions are the locally minimum and strongest ones. The correctness proof of the proposed algorithm is also included in the paper. An implemented tool, test data, and experimental results are presented and discussed.

Electronic payment protocols provide secure service for electronic commerce transactions and protect private information from malicious entities in a network. Formal methods have been introduced to verify the security of electronic payment protocols; however, these methods concentrate on the accountability and fairness of the protocols, without considering the impact caused by timeliness. To make up for this deficiency, we present a formal method to analyze the security properties of electronic payment protocols, namely, accountability, fairness and timeliness. We add a concise time expression to an existing logical reasoning method to represent the event time and extend the time characteristics of the logical inference rules. Then, the Netbill protocol is analyzed with our formal method, and we find that the fairness of the protocol is not satisfied due to the timeliness problem. The results illustrate that our formal method can analyze the key properties of electronic payment protocols. Furthermore, it can be used to verify the time properties of other security protocols.

Event-centered information integration is regarded as one of the most pressing issues in improving disaster emergency management. Ontology plays an increasingly important role in emergency information integration, and provides the possibility for emergency reasoning. However, the development of event ontology for disaster emergency is a laborious and difficult task due to the increasingly scale and complexity of emergencies. Ontology pattern is a modeling solution to solve the recurrent ontology design problem, which can improve the efficiency of ontology development by reusing patterns. By study on characteristics of numerous emergencies, this paper proposes a generic ontology pattern for emergency system modeling. Based on the emergency ontology pattern, a set of reasoning rules for emergency-evolution, emergency-solution and emergency-resource utilization reasoning were proposed to conduct emergency knowledge reasoning and q.

With the Semantic Web data standards defined, more applications demand inference engines in providing support for intelligent processing of the Semantic Web data. Rule-based inference engines or rule-based reasoners are used in many domains, such as in clinical support, and e-commerce recommender system development. This article reviews and compares key features of three freely-available rule-based reasoners: Jena inference engine, Euler YAP Engine, and BaseVISor. A performance evaluation study was conducted to assess the scalability and efficiency of these systems using data and rule sets adapted from the Berlin SPARQL Benchmark. We describe our methodology in assessing rule-based reasoners based on the benchmark. The study result shows the efficiency of the systems in performing reasoning tasks over different data sizes and rules involving various rule properties. The review and comparison results can provide a basis for users in choosing appropriate rule-based inference engines to match their application requirements.

In this letter, we propose a robust and fast tracking framework by combining local and global appearance models to cope with partial occlusion and pose variations. The global appearance model is represented by a correlation filter to efficiently estimate the movement of the target and the local appearance model is represented by local feature points to handle partial occlusion and scale variations. Then global and local appearance models are unified via the Bayesian inference in our tracking framework. We experimentally demonstrate the effectiveness of the proposed method in both terms of accuracy and time complexity, which takes 12ms per frame on average for benchmark datasets.

Recommender systems have attracted attention in both the academic and the business areas. They aim to give users more intelligent methods for navigating and identifying complex information spaces, especially in e-commerce domain. However, these systems still have to overcome certain limitations that reduce their performance, such as overspecialization of recommendations, cold-start, and difficulties when items with unequal probability distribution exist. A novel approach addresses the above issues through a case-based recommendation methodology which is a form of content-based recommendation that is well suited to many product recommendation domains, owing to the clear organization of users' needs and preferences. Unfortunately, the experience-based roots of case-based reasoning are not clearly reflected in case-based recommenders. In other words, the concept that product cases, which are usually fixed feature-based tuples, are experiential is not adopted well in case-based recommenders. To solve this problem as well as the recommenders' rating sparsity issue, one can use product reviews which are generated from users' experience with the product a basis of product information. Our approach adapts the use of sentiment scores along with feature similarity throughout the recommendation unlike traditional case-based recommender systems, which tend to depend entirely on pure similarity-based approaches. This paper models product cases with the products' features and sentiment scores at the feature level and product level. Thus, combining user experience and similarity measures improves the recommender performance and gives users more flexibility to choose whether they prefer products more similar to their query or better qualified products. We present the results using different evaluation methods for different case structures, different numbers of similar cases retrieved and multilevel sentiment-approaches. The recommender performance was highly improved with the use of feature-level sentiment approach, which recommends product cases that are similar to the query but favored for customers.

The amount of seasonings used during food preparation is quite important information for modern people to enable them to cook delicious dishes as well as to take care for their health. In this paper, we propose a near real-time automated system for measuring and recording the amount of seasonings used during food preparation. Our proposed system is equipped with two devices: electronic scales and a camera. Seasoning bottles are basically placed on the electronic scales in the proposed system, and the scales continually measure the total weight of the bottles placed on them. When a chef uses a certain seasoning, he/she first picks up the bottle containing it from the scales, then adds the seasoning to a dish, and then returns the bottle to the scales. In this process, the chef's picking and returning actions are monitored by the camera. The consumed amount of each seasoning is calculated as the difference in weight between before and after it is used. We evaluated the performance of the proposed system with experiments in 301 trials in actual food preparation performed by seven participants. The results revealed that our system successfully measured the consumption of seasonings in 60.1% of all the trials.

Prediction of seasonal infectious disease spread is traditionally dealt with as a function of time. Typical methods are time series analysis such as ARIMA (autoregressive, integrated, and moving average) or ANN (artificial neural networks). However, if we regard the time series data as the matrix form, e.g., consisting of yearly magnitude in row and weekly trend in column, we may expect to use a different method (matrix approach) to predict the disease spread when seasonality is dominant. The MD (matrix decomposition) method is the one method which is used in recommendation systems. The other is the IRT (item response theory) used in ability evaluation systems. In this paper, we apply these two methods to predict the disease spread in the case of infectious gastroenteritis caused by norovirus in Japan, and compare the results obtained by using two conventional methods in forecasting, ARIMA and ANN. We have found that the matrix approach is simple and useful in prediction for the seasonal infectious disease spread.

This paper reports on the trending literature of occlusion handling in the task of online visual tracking. The discussion first explores visual tracking realm and pinpoints the necessity of dedicated attention to the occlusion problem. The findings suggest that although occlusion detection facilitated tracking impressively, it has been largely ignored. The literature further showed that the mainstream of the research is gathered around human tracking and crowd analysis. This is followed by a novel taxonomy of types of occlusion and challenges arising from it, during and after the emergence of an occlusion. The discussion then focuses on an investigation of the approaches to handle the occlusion in the frame-by-frame basis. Literature analysis reveals that researchers examined every aspect of a tracker design that is hypothesized as beneficial in the robust tracking under occlusion. State-of-the-art solutions identified in the literature involved various camera settings, simplifying assumptions, appearance and motion models, target state representations and observation models. The identified clusters are then analyzed and discussed, and their merits and demerits are explained. Finally, areas of potential for future research are presented.

In non-destructive testing (NDT), ultrasonic echo is often an overlapping multi-echo signals with noise. However, the accurate estimation of ultrasonic time-of-flight (TOF) is essential in NDT. In this letter, a novel method for TOF estimation through envelope is proposed. Firstly, the wavelet denoising technique is applied to the noisy echo to improve the estimation accuracy. Then, the Hilbert transform (HT) is used in ultrasonic signal processing in order to extract the envelope of the echo. Finally, the TOF of each component of multi-echo signals is estimated by the local maximum point of signal envelope. Furthermore, the time resolution of time-overlapping ultrasonic echoes is discussed. Numerical simulation has been carried out to show the performances of the proposed method in estimating TOF of ultrasonic signal.

Earlier attempts to deploy two units of parametric loudspeakers have shown encouraging results in improving the accuracy of spatial audio reproductions. As compared to a pair of conventional loudspeakers, this improvement is mainly a result of being free of crosstalk due to the sharp directivity of the parametric loudspeaker. By replacing the normal parametric loudspeaker with the steerable parametric loudspeaker, a flexible sweet spot can be created that tolerates head movements of the listener. However, spatial aliasing effects of the primary frequency waves are always observed in the steerable parametric loudspeaker. We are motivated to make use of the spatial aliasing effects to create two sound beams from one unit of the steerable parametric loudspeaker. Hence, a reduction of power consumption and physical size can be achieved by cutting down the number of loudspeakers used in an audio system. By introducing a new parameter, namely the relative steering angle, we propose a stereophonic beamsteering method that can control the amplitude difference corresponding to the interaural level difference (ILD) between two sound beams. Currently, this proposed method does not support the reproduction of interaural time differences (ITD).

A two-stage classifier is proposed that identifies criminal charges and a range of punishments given a set of case facts and attributes. Our supervised-learning model focuses only on the offences against life and body section of the criminal law code of Thailand. The first stage identifies a set of diagnostic issues from the case facts using a set of artificial neural networks (ANNs) modularized in hierarchical order. The second stage extracts a set of legal elements from the diagnostic issues by employing a set of C4.5 decision tree classifiers. These linked modular networks of ANNs and decision trees form an effective system in terms of determining power and the ability to trace or infer the relevant legal reasoning behind the determination. Isolated and system-integrated experiments are conducted to measure the performance of the proposed system. The overall accuracy of the integrated system can exceed 90%. An actual case is also demonstrated to show the effectiveness of the proposed system.