In devices that are used for RFID tags and sensor networks, the number of available hardware resources is extremely small. Since there is a risk of information leakage from these devices, the data they handle must be enciphered. A lightweight block cipher must be used in these devices, due to their resource constraints. Although the encryption algorithms are theoretically safe, it has been recently reported that confidential information could be illegally revealed when the encryption algorithms are used in electronic circuits. In particular, fault analysis attacks have become extremely serious problems. Fault analysis attacks intentionally generate operation errors during the encryption processing and illegally obtain confidential information by pairing an incorrect cipher text and a correct cipher text. To secure the safety of lightweight block ciphers in the future, the tamper resistance against fault analysis attacks must be verified. The present study proposes a new fault analysis method for PRINCE. The proposed method introduces hierarchical attack model using statistical processing to reveal secret keys. Simulation results prove the validity of the proposed method.
In Japan, university students need to enhance their quality of learning. Particularly, it is necessary for university students to acquire “Academic Study Skills. ” The skills are a vital foundation for enhancing their quality of learning at university. These may include areas such as academic writing, critical thinking, learning study strategies, organizing their tasks, developing time management, exploring their learning styles, etc. In order to acquire the skills, university students must have an awareness of acquiring a proactive learning behavior. On the other hand, teachers should provide an opportunity to reflect and improve their learning behavior. That is to say, university students must shift from a passive learning behavior to a proactive learning behavior. In this paper, we give university students the opportunity to reflect and improve their learning behavior. One goal is to help them become aware of understanding the importance of academic study skills. We therefore focus on “the students' learning behavior,” and we visualize it as a queuing model. We express the queuing model as the VCHS (Various Customers, Heterogeneous Servers) queuing model with smarter servers. We then present the knowledge and graphs which were obtained by the performance evaluation of the model to the university students.
This paper describes an equivalent circuit of a photovoltaic system. The maximum power point is estimated with the equivalent circuit. A calculation method which estimates output voltage and current at the maximum power point by the use of the intensity of sunlight is introduced. The proposed method determines the starting voltage of the MPPT control of the PV system. Simulation results and preliminary experimental results show the feasibility of the proposed method.
We developed a climbing plan sharing system. While climbing plan documents that have been created in Microsoft Word or PDF format have a lot of climbing information, a computer cannot understand climbing information efficiently because the climbing information is written in a natural language. We developed a system that converts a climbing plan document into a machine-readable climbing plan. Also we have implemented a function to share climbing plans among users. The system can display a climbing plan on a Web browser effectively. Our system can help climbers to share machine-readable climbing plan documents. Therefore, users can share and accumulate climbing information that are difficult a climbing plan document written in a natural language.
Levitation caused the vertical vibration of a piezoelectric actuator is studied. Levitation mechanisms levitate using a squeeze effect generated by the vertical vibration. The levitation mechanisms usually levitate at the resonance frequency. However, if the vibration amplitude of the piezoelectric actuator at the resonance frequency is small, the levitation mechanism does not levitate. Therefore, the vibration amplitude affects the levitation height of the mechanism. In this paper, we describe the applied voltage, the load current, the power consumption, and the phase difference between the voltage and current.
This paper presents a real-time method that simulates motion of snowflakes on a windshield. A few minutes after snowflakes attach on a windshield, the snowflakes start to slip down. And they reach to the bottom of the windshield, by repeating the sliding and the stopping. During the sliding, snowflakes change motion direction at random. If a snowflake hits another, they together revolve and slip down. Our method simulates such motion with the particle model. In addition, snow on the windshield is gathered by a wiper. To represent this, a mesh is attached along the windshield wiper. The mesh and the wiper rotate together. When a snowflake contact to the wiper blade, it is stored to the cell of the mesh located at the contact point. If the cell is already filled, the snowflake is stored to the more outside cell. As the result, form of the snow stacked on the wiper is created. It is confirmed that our method can display similar movement of snow on the windshield in the comparison of video.
We proposed the method which generates ground texture by evaluating pressure applied at each point. The method standardizes diameter of clod to original size and its 1/2, 1/4, and 1/8 size, then defines condition of land by heft percentage of the clod of the four kinds of size. At each point of the ground, pressure caused by objects that pass through on the point is accumulated. Whenever the pressure exceeds the threshold, fixed ratio of weight of clod is shifted to its half size. As the result, at the place where many objects pass, heft percentage of smaller clod increases. Ground surface is defined as a multi-layered texture. It is created by combining four basic texture images. Each of the images is synthesized to correspond to the size of clod by the procedural method. From each of the texture images, round areas are selected and allocated to one of the layers at random. Diameter of the area is equal to size of the clod. And the numbers of the area is proportionate to the heft percentage of the clod. The multilayer texture is combined to one layer, after adding shade and shadow. And finally, it is mapped on the ground.
Extraction of facial feature points is essential task for many kinds of applications of face images. The feasibility of facial features extraction is determined by not only its accuracy but processing time. Some applications require real-time detection of facial features. This research aims to propose a method of facial features extraction by an accelerated implementation of circular Hough transform with gradients and appearance evaluation by histogram of gradient features. Experiment using FERET database shows that the proposed method successfully extracted eyes, nose and mouth for 98.44%, 99.50% and 98.79% of frontal face images in the dataset.
We propose a review supporting system of trial lessons with image processing techniques. Candidates of a high/junior schoolteacher practice in the trial lesson. A subjective evaluation from the other candidates is used as a feedback of the lesson. However, the quality of the feedback result depends on the skill and experience of the evaluators. Review using a recorded video of the lesson can improve the skill of the candidate, however this method requires a long time. Therefore, we focus on an objective evaluation from a recorded video, in which the behavior is visualized in static images. The system provides the three processes, visualization of teacher movement, extraction of writing on blackboard, and blind spot alert. Our implemental approach is to calculate an optical flow indicating human movement, and to draw segment for visualization. An exclusive OR between previous and target frames is calculated for detecting a region of writing. The blind spot by the candidate is estimated with the human position and the region of the writing, and an alert is provided as visual information in the timing. The accuracy of the alert timing are about 80% of the precision and recall with the author visual judgment. In addition, some positive comments are obtained in a questionnaire result, and the availability is indicated in this paper.
When people listen to other's speech for the first time, they always attribute personality traits to the speaker subconsciously. We consider that if robots can predict personality traits of users from their speech, the communication in Human-Machine Interaction will improve significantly. This paper proposes an approach for the automatic estimation of the traits, in which the listeners attribute to unanimous speakers. And the discrimination experiments based on Hidden Markov Model (HMM) and Canonical Discrimination Analysis (CDA) show that, it is possible to predict with high accuracy (more than 75 percent), whether a speaker is perceived to be in the higher or lower part of the “Extraversion”, “Openness” and “conscientiousness” by using Non-verbal information.
We have introduced pair work to university-level computer literacy education as opposed to the traditional one, standalone teaching. Since 2009, we have performed a survey to establish a pair formation method that optimizes student combinations. Thus far, we have clarified that during pair formation, gender, a gap in basic academic ability, and personality characteristics are more effective indicators compared with typing speed, interest in computers, and computer experience. In this paper, based on the previous research results of more than five years, we propose a new pair combination method that uses three kinds of information which are gender, basic academic ability and personality characteristics, called GAP (Gender, basic Academic ability, and Personality) Method. In 2014, we surveyed for new pair-combination using PS (Personality Score) which has shown personality characteristics and conducted verification experiments for GAP Method in 2015. As a result, it became clear that the pair combination method using GAP Method is higher pair effect than random combined pairs.
In modern VLSI designs, a flip-chip package is widely used to meet the higher integration density and the larger I/O counts of circuits. Recently the I/O buffers are mapped onto bump balls without changing the module placement using re-distribution layer (RDL) in flip-chip designs. In this research, a sorting-based I/O connection assignment and non-Manhattan RDL routing method is proposed for area I/O flip-chip designs. The approach initially assigns the I/O buffers to bump balls by sorting the Manhattan distance between them. Three kinds of pair-exchange procedures are then carried out to improve the initial assignment. Then to shorten the wire length, non-Manhattan RDL routing is adopted to connect the I/O buffers and bump balls. Finally some un-routed connections are ripped up and rerouted. The experimental results show that the proposed method is able to obtain the routes with shorter wire length in reasonable CPU times.
In this paper, we propose an AR interface that enables additional manipulation of displayed objects via finger motions while displaying the object based on palm posture. The palm posture is estimated by the proposed AR interface based on the feature-points that are robust against the variation of hand silhouettes depending on finger motion. By utilizing the difference of hand silhouettes depending on the motion of five fingers, we provide the additional manipulations of a displayed object for the AR interface. We evaluate the drawing accuracy and processing time of the proposed interface under the condition that a hand includes successive changes of its posture and finger motion. Additionally, to confirm the feasibility of the proposed AR interface, we evaluated the manipulation with two prototype applications that utilize the difference of hand silhouettes depending on finger motion. The experimental results show that the proposed AR interface has the abilities to switch objects displayed on a palm and to change the transparency of a displayed object.
In Japan, we frequently bow to various people in daily life. Therefore, we should acquire the polite ritsurei so that we don't give the other party the unpleasant. The goal of the research is development of ritsurei teaching system which is composed of ritsurei evaluation phase and user teaching phase. The ritsurei evaluation phase evaluates ritsurei based on position of human body as back of the head, the back, and the waist. However, it is difficult to track these positions. Because back of the head, the back, and waist have little characteristic pattern. In this paper, we estimate bowing angle by tracking relative movements of local features on human body contour. Experimental results have shown that the proposed method has increased estimation accuracy of angle by 40% compared with the system which tracks back of the head, the back, and the waist without contour of human body. In addition, identification accuracy of proposed system has been equal to identification accuracy of marker tracking system.
In a dental treatment, doctors and patients often feel strong discomfort feeling by hearing the dental treatment sound. It is caused by spectral peaks in a higher frequency. For reducing discomfort feeling of spectral peaks, we have previously proposed the discomfort reduction method that can mask spectral peaks on basis of auditory masking by emitting the masking signal. In principle, this method has a problem that it requires emission of the masking signal with higher power. It may inflict discomfort feeling on listeners. On the other hand, we focus on chords in music theory because they are generally known as typical comfortable sounds. The chords have the spectral structure that multiple spectral peaks are held with specific rules, and the temporal structure changes on basis of the chord progression of musical pieces. These suggest that we may design the comfortable sound by reforming the spectral structure of the dental treatment sound to that of chords. In this paper, we therefore propose the comfortable sound design method based on chord-forming of the spectral peak in time and frequency domains for the dental treatment sound. As a result of subjective evaluation experiments, we confirmed the effectiveness of the proposed method.
In action recognition, in order to obtain high performance classifiers, it is necessary to feed the training algorithms enough labeled data. Since labeling is a very expensive task, it is important to develop approaches which can efficiently reuse existing labeled data. In this work, we consider the task of utilizing labeled data from one dataset (source dataset) to train action classifiers for data from another completely unlabeled dataset (target dataset). We propose a novel approach for such a task by extending the well-known self-training algorithm to including data selection or feature selection. The superior of our approach has been confirmed by benchmark datasets.
The paper presents a model to identify person's actions using non-portable vibration sensor. In Japan, because solitary aged persons, who require the care, rapidly increases, a watching support system is needed to protect such aged persons. Vibrations generated by person's action are monitored by a non-portable vibration sensor. Some indices to evaluate the degree of similarity among different monitoring vibration data are formulated and are integrated based on the principal component analysis. Proposed similarity judgment method applies to an actual wooden 2-story house and discusses the judgment accuracy. As a result, it is shown that the purposed model enables us to accurately identify 15 person's action modes using a single non-portable vibration sensor.
This paper describes a matrix circuit converting a four-element multi-beam antenna to a six-element antenna. Signal outputs from the four-element antenna are distributed to six outputs by Binomial Directivity. This circuit is designed at 2.45GHz and uses micro strip lines and chip resistors. Its characteristics are evaluated by RF circuit simulation. The circuit is fabricated and its characteristics are measured with a microwave network analyzer. Experimental results show that the measured value agreed well with the simulated value, and we were able to produce a valid circuit.
Eye fixation and its time were measured when pairs of food pictures were displayed; left side picture is for reference, and right side picture is for subjective evaluation by the stimulus-comparison method with opinions ‘Like’, ‘Same’, and ‘Dislike’. Eye fixation time ratio FTR was obtained as a ratio of the fixation time for evaluation pictures to that for reference ones. Results showed FTR in the case of ‘Like’ was larger.
In this letter, we propose a new multi-objective optimization method based on Artificial Bee Colony Algorithm. The proposed method takes advantage of division of search functions that characterize Artificial Bee Colony Algorithm. The performance of the proposed method is verified by numerical simulation using three typical benchmark problems.