Search

Applying image analysis techniques using deep learning in the consulting field is a new and promising field. However, there are no complete studies on it yet. This study mentions image recognition related literatures, summary of cases of artificial intelligence use in counseling field.

The hearing aids transmit voice in the manner of transmitting voice by air conduction method which is the same generation mechanism as ambient noise and various voice signals. The problem is that it is difficult to transmit clear speech in robust environmental noise, and it is difficult to analyze a signal in a signal that is similar to a voice signal, which is a desired transmission signal. Recently, in order to overcome these disadvantages, a method of transmitting voice signals using a bone conduction system having different mechanisms of occurrence has been used. However, the effectiveness of the bone conduction method and the transmission path of the voice signal transmission are not clarified, and thus there is a problem with the verification method of the efficiency. In this paper, to overcome these problems, we verified the signal transmission efficiency of an animal model with bone conduction oscillator using ABR (Auditory Brainstem Response) and MeMRI (Manganese Enhancement MRI).

Insect identification has the disadvantage that it is difficult for non-experts to carry out due to the specificity of insects. Therefore, it is necessary for the general user to use auxiliary tools such as books to identify insects for education such as ecological learning. In recent years, researches using Deep Learning in fields such as object detection, behavior recognition, voice recognition as well as cancer detection in medical field have been actively conducted and show excellent results. In this paper, we developed a classification application that can be used in mobile phones with high automation and portability to solve the above insect classification problems. Experiments were conducted on 30 insect species selected for observable insects irrespective of environmental factors such as habitat and season, and the transform learning were applied to ResNet, which showed excellent performance in ILSVRC to classify forest insect. Our system achieved an average insect classification accuracy of 94%, an insect classification speed of 0.03 sec, and an insect photo transmission of 0.5 sec to output this information. This has proven that non-experts provide the appropriate performance to use.

Research on the artificial intelligence is increasing with the improvement of computing power and the development of algorithm theory. In particular, the deep neural network, which is a field of machine learning, is widely used in artificial intelligence because it can process data that cannot be solved by conventional shallow neural networks more effectively. Implementation of a deep neural network is generally based on popularized neural networks with excellent generalization performance, which saves time and effort. However, it is difficult to guess which deep neural networks and optimization methods can achieve the best performance in their dataset. In this paper, we have tested the four convolutional neural networks (ConvNet) and four weight update methods. Experiments were conducted using a 5-fold cross-validation based on insect image dataset. As a result, the ResNet-50 and AdaDelta combination showed the best performance (89.98 ± 1.40)% in the insect dataset.

Along with the popularity and widespread use of automated teller machines (ATMs), ATM frauds are also increasing drastically these days. Shoulder-surfing attacks, such as card skimming, PIN capturing using fake machines or fake PIN pads, are the most common methods used by adversaries to capture data from the magnetic stripe on the back of the ATM card. The main problem lies in the existing static PIN-based authentication mechanism which does not provide any security measure (besides displaying asterisks when an user enters a preassigned PIN to the ATM). In this paper, we give a novel approach called DynamicPIN for secure ATM authentication, which is resilient to shoulder-surfing attacks. DynamicPIN is very simple, does not require any hardware changes, and does not pose any significant overhead to the system. A realtime experimental study showed that DynamicPIN improves significantly the ATM authentication compared to the existing static PIN-based authentication mechanism.

It is necessary to develop an automated insect classification method in order to create high added value of insects. Recently, image classification work applying deep learning shows the best result of classification performance. In this paper, we use convolutional neural network to classify insects and examine the effect of the classification performance with the varying number of kernels in the convolution layer and with the different image datasets. Experiments were conducted on 27 insect classes of ImageNet dataset with AlexNet. We confirmed that the classification performance of neural network improves where insects are well represented at the center of the image, and when the number of convolution kernels increases.

Recent maritime trading volume has been growing continuously and being activated marine leisure market. The government has to control the sea traffic volume through the facility operations such as maritime traffic centers. But more complex marine traffic situation in these trends, according to trend are marine accidents continue to occur. Marine accidents have a difficult characteristic to recover when if suffered environmental and human damage. There is also a trend, scale of the accident has been increased in size. High-risk vessels (such as Cruise ships, Tankers) are required monitoring and analysis of the detailed situation than the general ship. There have been a variety of research in order to solve the problem. However, the data elements are reflected in the shore with the current maritime are only limited information about AIS, Radar information. We are implementing IoT sensor interfacing system capable of transmitting various ship sensing data to the shore-side system and also proposing a Safe Navigation Status Analysis System utilizing the collected data.

Due to raised incidences of stroke, paralysis, or other diseases along with dramatic increment of life expectancy, the number of patients with movement disability has been increasing continuously. Repetitive and intensive voluntary movements in physical therapy are important factors that facilitate significant improvement for motor-impaired patients. The emergence of rehabilitation robotic devices has stimulated the development of physical therapy. However, most of current robotic devices for upper limb are poor in user-friendly interface and bulky as well as assisting only limited part(s) of arm. We proposed Smart Portable Rehabilitation Exoskeletal Device (SPRED), which is a portable, tele-operatable, and effective exoskeleton type of upper limb rehabilitation robotic device controlled by multimodal signals with smart interfaces for both patients and therapists. The SPRED system supports full range of joint movements and assists disabled arms more naturally through highly accurate, adaptable, and fast responses based on muscle strength, brain activity, and motion tracking technology. The compact size and wireless device allows patients to carry the device during their daily activities so that they can naturally lengthen the training duration and conclude more effective clinical results eventually. We believe that the research will contribute to development a new generation of exoskeleton type of rehabilitation robotic device for upper limb. As a first step towards the proposed system, this paper presents the design of SPRED and the mirroring motion based self-tuning concept is illustrated. Its simulation result demonstrates its potential in upper limb rehabilitation.

This study examines the neural basis of long-term memory formation. Subjects are instructed to learn association pairs on a given day and their retention performance is recorded on the next. We perform statistical analysis as well as single-trial classification to investigate whether EEG is able to predict longterm memory formation in single trials. Our preliminary results confirm previous neurophysiological findings, and show that the prediction of long-term memory formation is possible.

In particular, social network service has changed the traditional way of communication between consumers and tourism enterprises, making them to collect information via itself. According to this situation, this thesis takes the users of social network sites as the objects for research, the technology acceptance model as theoretical basis, tourists cognitive trust and emotional trust as intermediate variables, and combines with the usefulness, dependence, openness and community conversation of social network service to build a hypothetical model, and then uses empirical research methods to explore the impact of social network service on tourists purchase intention. The empirical results show that the dependence and community conversation will have a direct positive impact on tourists purchase intention. The results prove the usefulness, dependence, openness, community conversation will have a positive impact on tourists purchase intention. Finally, according to the conclusions this thesis puts forward feasible marketing suggestions for tourism enterprises on how to use social network service, and then describes the outlook for future research.

The recent rise in maritime traffic volume, resulted from an increase in international marine trading volumes and the growing popularity of marine leisure activities, has increased the frequency of marine accidents. As human errors have accounted for the greatest percentage of marine accidents, a series of navigation assistance systems capable of assisting mates have been introduced. However, vessel data, which can be found in the previously proposed studies, are merely composed of a number of fixed sets of analyzed and assumed data, without mentioning any real-time data management techniques that can be embodied as real-time systems. In addition, as the proposed vessel data processing techniques are designed for Vessel Traffic Service Center, it is difficult to find data management techniques exclusively designed in consideration of the viewpoint of vessel operators. In this regard, a real-time vessel data management technique targeted at vessels operating at sea is to be introduced by this paper.

The recent rise in maritime traffic volume, resulted from an increase in international marine trading volumes and the growing popularity of marine leisure activities, has increased the frequency of marine accidents. The majority of maritime accidents are caused by humans account, and a diversity of assistance systems that can assist officers including the technologies for detecting abnormal navigating ships have been proposed to reduce maritime accidents caused by human factors. However, as previous collision risk assessment methods only take into account collision risks at a single point in time, there may be high-risk vessels regarded "not risky" due to erroneous data generated from malfunctioning devices. In this paper, we proposed a new collision risk assessment systems, systems for identifying abnormal navigating ship, and cumulative collision risk assessment systems using fuzzy theory, where such systems are integrated in a single system.

Vessels exhibiting abnormal navigation patterns (e.g., weaving in and out of courses or rotating in the same position) may have a serious impact on other vessels staying on normal courses. For this reason, ground VTS centers are keeping track of criminal vessels or damaged vessels in tandem with marine police. However, the number of available studies aimed at assisting the identification of seemingly apparent risk factors resulting from human errors has been next to nothing to date. In this respect, this study intends to implement an intelligent system that can identify vessels exhibiting abnormal navigation patterns based on fuzzy inference, in order to assist controllers and mates alike.

Face detection and recognition have become more and more popular, especially in the era of hand-held devices. As a result, many algorithms have been developed to process face images. However, many of those also have problems with uneven illumination effects, because images have been captured under various lighting conditions. In this paper, we introduce a heuristic approach for shadow and light regions fast detection in face images. The results will be used as clues for other correction algorithms. Within the available samples of the face region, we use the K-means algorithm to cluster pixels into shadow, light and light-balanced regions. Since the heuristic K-means method may generate misclassified pixels, we use image processing techniques to enhance the clustered results. Experiments conducted on the Caltech face dataset show that our proposed approach can robustly, totally and quickly detect shadow and light regions in face images.

Cloud computing promises to reshape the way IT service is produced and consumed by virtualizing computing resources. As if cloud providers offer virtualized CPU and storage resources via machine virtualization and distributed storage technologies respectively, the providers wish to offer to their customers' virtual networks running on the cloud. With cloud-based virtual networks (CVNs), cloud customers (often, corporate customers) can easily build new sites, effectively expanding their enterprise networks into the cloud and thus leveraging the entire benefits of cloud computing (agility, manageability, low cost, etc.). In this paper, we identify core requirements and objectives for managing CVNs. We then define a high-level autonomic CVN-management framework by demarcating between the customer and provider with CVN Management Interface (CMI). Finally we propose an open-flow based autonomic cloud data center management solution based on CVN management framework.

This paper is proposed a novel method to approximate the maximum power for a photovoltaic inverter system and tracking method. It is designed for power systems application and utilities. The proposed Maximum Power Point Tracking(MPPT) control has the advantage to provide a new simple way to approximate the optimal or rated voltage, the optimal or rated current and maximum power rating produced by a solar panel and the photovoltaic inverter. And this straightforward method will be named linear reoriented coordinates method(LRCM) with the advantage that Pmax and can be approximated using the same variable as the dynamic model without using complicate approximations or Taylor series. Furthermore tracking method is improved over 50% photovoltaic efficiency. This paper is proposed MPPT using LRMC and tracking method using weather condition of domestic moderate program technique. This paper is proposed the experimental results to verify the effectiveness of the new methods.

In this paper proposed the solar tracking system to use a fuzzy based on PC in order to increase an output of the PV array. The solar tracking system operated two DC motors driving by signal of photo sensor. The control of dual axes is not an easy task due to nonlinear dynamics and unavailability of the parameters. Recently, artificial intelligent control of the fuzzy control, neural-network and genetic algorithm etc. have been studied. The fuzzy control made a nonlinear dynamics to well perform and had a robust and highly efficient characteristic about a parameter variable as well as a nonlinear characteristic. Hence the fuzzy control was used to perform the tracking system after comparing with error values of setting-up, nonlinear altitude and azimuth. In this paper designed a fuzzy controller for improving output of PV array and evaluated comparison with efficient of conventional P controller. The data which were obtained by experiment were able to show a validity of the proposed controller.

Conducting polymer (Polypyrrole) and tinoxide (SnO2) composite films have been fabricated with layer-by-layer technique. SnO2 film was screen-printed on Al2O3 substrate and then it was dip-coated with polypyrrole (PPy) solution. The microstructures of composite films were evaluated by a field emission scanning electron microscope (FE-SEM). It was observed the sensing properties of PPy/SnO2 sensor to various VOCs. Methanol, ethanol, benzene and toluene were analyzed. The target vapors were vaporized using their vapor presents and their concentrations were precisely controlled using mass flow controller (MFC) and temperature controller. The reproducibility of sensing properties was evaluated and the humidity effect of PPy/SnO2 was checked. This electronic document is a "live" template. The various components of your paper [title, text, heads, etc.] are already defined on the style sheet, as illustrated by the portions given in this document.

Nanowires of conducting polypyrrole (PPy) were synthesized by chemical polymerization methode. The 0.18 mol of pyrrole (Py) as monomer, 0.18 mol of dodecyl-benzene sulfonic acid (DBSA) as dopant and 0.26 mol of anhydrous iron(lll) tri chloride (FeCl3) as oxidant were mixed and stirred in 200 ml of distilled water at 0degC. The reaction was carried out for three different times (6 h, 24 h, 40 h) and reaction rate was 200 rpm. The morphology of PPy was evaluated by field emission scanning electron microscopy (FE-SEM). It was found that the morphology of nanowire PPy was affected by reaction time and aggregation rate. When the polymerization time was getting longer, nanowire structure was appeared clearly. The sensitivities of nanowire PPy were investigated with volatile organic compounds (VOCs).

The gigabit Ethernet switch is already being deployed in large numbers in both corporate and public data networks, and has begun to move Ethernet from the region of the local area network out to encompass the metro area network. The main issue for 10-gigabit Ethernet switch development is the need of a new network processor that ern process high-speed packets forwarding on line interface module. In this paper, we compare functionality of the commercial NPUs. And we introduce architecture of the 10-gigabit Ethernet backbone switch supporting L2 and L3 switching functions.

We have proposed the structure of an interworking system and the method of inter processor communication for providing existing narrowband network interworking in ATM switching system. The proposed interworking module consists of 8 low speed application specific units (LASU), 3 high speed application specific units (HASU) and a interworking control processor (IWP). The LASU and the HASU are application specific and the IWP is application independent. From the interworking point of view, application specific means non-ATM function such as DS1/E1/DS3 frame relay service and circuit emulation service and application independent means common ATM function. The inter processor communication function is to transfer call control messages, OAM messages, TDM channel establishment messages between the main processor and interworking processor.