The TFT-LCD image has non-uniform brightness that is the major difficulty of finding the visible defect called Mura in the field. To facilitate Mura detection, background signal shading should level off and Mura signal must be amplified. In this paper, Mura signal amplification and background signal flattening method is proposed based on human visual system (HVS). The proposed DC normalized contrast sensitivity function (CSF) is used for the Mura signal amplification and polynomial regression (PR) is used to level off the background signal. In the enhanced image, tri-modal thresholding segmentation technique is used for finding Dark and White Mura at the same time. To select reliable defect, falsely detected invisible region is eliminated based on Weber's Law. By the experimental results of artificially generated 1-d signal and TFT-LCD image, proposed algorithm has novel enhancement results and can be applied to real automated inspection system.

Service robots need to be able to recognize and identify objects located within complex backgrounds. Since no single method may work in every situation, several methods need to be combined and robots have to select the appropriate one automatically. In this paper we propose a scheme to classify situations depending on the characteristics of the object of interest and user demand. We classify situations into four groups and employ different techniques for each. We use Scale-invariant feature transform (SIFT), Kernel Principal Components Analysis (KPCA) in conjunction with Support Vector Machine (SVM) using intensity, color, and Gabor features for five object categories. We show that the use of appropriate features is important for the use of KPCA and SVM based techniques on different kinds of objects. Through experiments we show that by using our categorization scheme a service robot can select an appropriate feature and method, and considerably improve its recognition performance. Yet, recognition is not perfect. Thus, we propose to combine the autonomous method with an interactive method that allows the robot to recognize the user request for a specific object and class when the robot fails to recognize the object. We also propose an interactive way to update the object model that is used to recognize an object upon failure in conjunction with the user's feedback.

Quantifying user satisfaction is essential, because the results can help service providers deliver better services. In this work, we propose a generalizable methodology, based on survival analysis, to quantify user satisfaction in terms of session times, i.e., the length of time users stay with an application. Unlike subjective human surveys, our methodology is based solely on passive measurement, which is more cost-efficient and better able to capture subconscious reactions. Furthermore, by using session times, rather than a specific performance indicator, such as the level of distortion of voice signals, the effects of other factors like loudness and sidetone, can also be captured by the developed models. Like survival analysis, our methodology is characterized by low complexity and a simple model-developing process. The feasibility of our methodology is demonstrated through case studies of ShenZhou Online, a commercial MMORPG in Taiwan, and the most prevalent VoIP application in the world, namely Skype. Through the model development process, we can also identify the most significant performance factors and their impacts on user satisfaction and discuss how they can be exploited to improve user experience and optimize resource allocation.

This paper proposes a robust omnidirectional audio-visual (AV) talker localizer for AV applications. The proposed localizer consists of two innovations. One of them is robust omnidirectional audio and visual features. The direction of arrival (DOA) estimation using an equilateral triangular microphone array, and human position estimation using an omnidirectional video camera extract the AV features. The other is a dynamic fusion of the AV features. The validity criterion, called the audio- or visual-localization counter, validates each audio- or visual-feature. The reliability criterion, called the speech arriving evaluator, acts as a dynamic weight to eliminate any prior statistical properties from its fusion procedure. The proposed localizer can compatibly achieve talker localization in a speech activity and user localization in a non-speech activity under the identical fusion rule. Talker localization experiments were conducted in an actual room to evaluate the effectiveness of the proposed localizer. The results confirmed that the talker localization performance of the proposed AV localizer using the validity and reliability criteria is superior to that of conventional localizers.

The thin film transistor liquid crystal display (TFT-LCD) image has nonuniform brightness, which is a major difficulty in finding the Mura defect region. To facilitate Mura segmentation, globally widely varying background signal must be flattened and then Mura signal must be enhanced. In this paper, Mura signal enhancement and background-signal-flattening methods using wavelet coefficient processing are proposed. The wavelet approximation coefficients are used for background-signal flattening, while wavelet detail coefficients are employed to magnify the Mura signal on the basis of an adapted contrast sensitivity function (CSF). Then, for the enhanced image, trimodal thresholding segmentation technique and a false-region elimination method based on the human visual system (HVS) are employed for reliable Mura segmentation. The experimental results show that the proposed algorithms produce promising results and can be applied to automated inspection systems for finding Muras in a TFT-LCD image.

In this paper, we describe a new ultrasonic oscillosensor and its application in a biological information measurement system. This ultrasonic sensor has a cylindrical tank of 26 mm (diameter)20 mm (height) filled with water and an ultrasonic probe. It detects the vibration of the target object by obtaining echo signals reflected from the water surface. This sensor can noninvasively detect the vibration of a patient by placing it under a bed frame. We propose a recognition system for humans in bed. Using this sensor, we could determine whether or not a patient is in the bed. Moreover, we propose a heart rate monitoring system using this sensor. When our system was tested on four volunteers, we successfully detected a heart rate comparable to that in the case of using an electrocardiograph. Fuzzy logic plays a primary role in the recognition. Consequently, this system can noninvasively determine whether a patient is in the bed as well as their heart rate using a constraint-free and compact device.

This paper proposes to use eye movements to characterize the performance of individuals in reviewing software documents. We design and implement a system called DRESREM, which measures and records eye movements of document reviewers. Based on the eye movements captured by eye tracking device, the system computes the line number of the document that the reviewer is currently looking at. The system can also record and play back how the eyes moved during the review process. To evaluate the effectiveness of the system we conducted an experiment to analyze 30 processes of source code review (6 programs, 5 subjects) using the system. As a result, we have identified a particular pattern, called scan, in the subject's eye movements. Quantitative analysis showed that reviewers who did not spend enough time on the scan took more time to find defects on average.

Machine learning techniques have long been applied in many fields and have gained a lot of success. The purpose of learning processes is generally to obtain a set of parameters based on a given data set by minimizing a certain objective function which can explain the data set in a maximum likelihood or minimum estimation error sense. However, most of the learned parameters are highly data dependent and rarely reflect the true physical mechanism that is involved in the observation data. In order to obtain the inherent knowledge involved in the observed data, it is necessary to combine physical models with learning process rather than only fitting the observations with a black box model. To reveal underlying properties of human speech production, we proposed a learning process based on a physiological articulatory model and a coarticulation model, where both of the models are derived from human mechanisms. A two-layer learning framework was designed to learn the parameters concerned with physiological level using the physiological articulatory model and the parameters in the motor planning level using the coarticulation model. The learning process was carried out on an articulatory database of human speech production. The learned parameters were evaluated by numerical experiments and listening tests. The phonetic targets obtained in the planning stage provided an evidence for understanding the virtual targets of human speech production. As a result, the model based learning process reveals the inherent mechanism of the human speech via the learned parameters with certain physical meaning.

This paper proposes a design method for representing monochrome medical X-ray images on an electronic display. The required quantizing resolution of the input density and output voltage are theoretically clarified. The proposed method makes it easier to determine the required quantizing resolution which is important in a X-ray diagnostic system.

In this paper, a new subspace-based speech enhancement algorithm is presented. First, we construct a perceptual filterbank from psycho-acoustic model and incorporate it in the subspace-based enhancement approach. This filterbank is created through a five-level wavelet packet decomposition. The masking properties of the human auditory system are then derived based on the perceptual filterbank. Finally, the prior SNR and the masking threshold of each critical band are taken to decide the attenuation factor of the optimal linear estimator. Five different types of in-car noises in TAICAR database were used in our evaluation. The experimental results demonstrated that our approach outperformed conventional subspace and spectral subtraction methods.

A communication means is presented for patients with amyotrophic lateral sclerosis in totally locked-in state who are completely unable to move any part of the body and have no usual communication means. The method utilizes changes in cerebral blood volume accompanied with changes in brain activity. When a patient is asked a question and the answer to it is 'yes', the patient makes his or her brain active. The change in blood volume at the frontal lobe is detected with near-infrared light. The instantaneous amplitude and phase of the change are calculated, and the maximum amplitude and phase change are obtained. The answer 'yes' or 'no' of the patient is detected using a discriminant analysis with these two quantities as variables. The rate of correct detection is 80% on average.

Existing vision substitute systems have insufficient spatial resolution to provide environmental information. To present detailed spatial information, we propose two stimulation methods to enhance transfer information using a 2-D tactile stimulator array. First, stimulators are divided into several groups. Since each stimulator group is activated alternately, the interval of stimulations can be shortened to less than the two-point discrimination threshold. In the case that stimulators are divided into two and four groups, the number of stimulators increases to twice and four times, respectively, that in the case of the two-point discrimination threshold. Further, a user selects the measurement range and the system presents targets within the range. The user acquires spatial information of the entire measurement area by changing the measurement range. This method can accurately present a range of targets. We examine and confirm these methods experimentally.

In human-computer interaction, Fitts' law has been applied in one-dimensional pointing task evaluation for some decades, and the usage of effective target width (We) in Fitts' law has been accepted as an international standard in ISO standards 9241-9 [4]. However, the discussion on the concrete methods for calculating We has not been developed comprehensively nor have the different methods of calculation been integrated. Therefore, this paper focuses on a detailed description and a comparison of the two main We calculation methods. One method is mapping all the abscissa data in one united relative coordinate system to perform the calculation (called CC method) and the other is dividing the data into two groups and mapping them in two separate coordinate systems (called SC method). We tested the accuracy of each method and compared both methods in a highly controlled experiment. The experiments' results and data analysis show that the CC method is better than the SC method for human computer interface modeling. These results will be instrumental for future application of Fitts' law.

Fitts' law has been applied in many studies to evaluate pointing tasks. However, the quantitative effect of using color in the interfaces has not been discussed in the literature. This paper introduces research on the effects of color in pointing tasks using Fitts' law as the evaluation method. Different colors and color presentation styles are applied in the experiments which are similar in design to the paradigmatic Fitts' law pointing task. The experimental results show that when the subjects use a mouse as the input device, there is no significant difference between an interface with a colored target and one with a white target in the mean performance time. The results also reveal that color presentation styles will offer no significant difference to pointing tasks when the mouse is applied. However, when the interface of tablet PC and pen was applied, subjects without much experience in tablet personal computer usage needed more time to perform the task with colored targets than with a white target. Furthermore, when the colors are changed randomly during the selection process, the difference is even more obvious. These results are confirmed by a Checking Experiment and a Learning Effect Experiment which we performed on different groups of subjects.

Adaptation process of retina helps human visual system to see a high dynamic range scene in real world. This paper presents a simple static local adaptation method for high dynamic range image compression based on a retinal model. The proposed simple model aims at recreating the same sensations between the real scene and the range compressed image on display device when viewed after reaching steady state local adaptation respectively. Our new model takes the display adaptation into account in relation to the scene adaptation based on the retinal model. In computing local adaptation, the use of nonlinear edge preserving bilateral filter presents a better tonal rendition in preserving the local contrast and details while avoiding banding artifacts normally seen in local methods. Finally, we demonstrate the effectiveness of the proposed model by estimating the color difference between the recreated image and the target visual image obtained by trial and error method.

As a result of the growth of sensor-enabled mobile devices, in recent years, users can utilize diverse digital contents everywhere and anytime. However, the interfaces of mobile applications are often unnatural due to limited computational capability, miniaturized input/output controls, and so on. To complement the poor user interface (UI) and fully utilize mobility as feature of mobile devices, we explore possibilities for a new UI of mobile devices. This paper describes the method for recognizing and analyzing a user's continuous action including the user's various gestures and postures. The application example we created is mobile game called AM-Fishing game on mobile devices that employ the accelerometer as the main interaction modality. The demonstration shows the evaluation for the system usability.

Pneumatic pressure, which is easy enough to be handled in comparison with hydraulic pressure and is endowed with high safety, is available for a power source of a robot arm to be utilized in concert with human beings to do various types of work. But pneumatic pressure is so low in comparison with hydraulic pressure that an air cylinder having a diameter long enough and stroke wide enough is required to obtain great output power. In this study, therefore, the investigation was made with layout of air cylinders and transmission mechanisms of the motion power directed toward the driving joints to be followed by development of a new humanoid robot arm with seven degrees of freedom in which air cylinders are compactly incorporated. To be concrete with this, contrivance was made with an endoskeleton structure allowing almost all of the structure materials of the individual arm joints to be shared by the air cylinder with incorporation of the air cylinder in the axes of the upper arm joint and forearm joints by paying attention to the fact that the cylinder itself has high strength. The evaluation experiments driving the robot arm referred to above were conducted by means of I-PD control. The results suggested that the mechanism of the robot with seven degrees of freedom having pneumatic actuators proposed in this study is useful as the humanoid robot arm. The quick and accurate motions were accomplished with I-PD control which is relatively easy to be dealt with but not suitable for non-linear actuator system.

An alternative human interface enabling the handicapped with severe motor disabilities to control an assistive system is presented. Since this interface relies on the biosignals originating from the contraction of muscles on the face during particular movements, even individuals with a paralyzed limb can use it with ease. For real-world application, a dedicated hardware module employing a general-purpose DSP was implemented and its validity tested on an electrically powered wheelchair. Furthermore, an additional attempt to reduce error rates to a minimum for stable operation was also made based on the entropy information inherent in the signals during the classification phase. In the experiments in which 11 subjects participated, it was found most of them could control the target system at their own will, and thus the proposed interface could be considered a potential alternative for the interaction of the severely handicapped with electronic systems.

An effective human-robot interaction is essential for wide penetration of service robots into the market. Such robot needs a vision system to recognize objects. It is, however, difficult to realize vision systems that can work in various conditions. More robust techniques of object recognition and image segmentation are essential. Thus, we have proposed to use the human user's assistance for objects recognition through speech. This paper presents a system that recognizes objects in occlusion and/or multicolor cases using geometric and photometric analysis of images. Based on the analysis results, the system makes a hypothesis of the scene. Then, it asks the user for confirmation by describing the hypothesis. If the hypothesis is not correct, the system generates another hypothesis until it correctly understands the scene. Through experiments on a real mobile robot, we have confirmed the usefulness of the system.

In this paper, we present a method for recognition of human activity as a series of actions from an image sequence. The difficulty with the problem is that there is a chicken-egg dilemma that each action needs to be extracted in advance for its recognition but the precise extraction is only possible after the action is correctly identified. In order to solve this dilemma, we use as many models as actions of our interest, and test each model against a given sequence to find a matched model for each action occurring in the sequence. For each action, a model is designed so as to represent any activity containing the action. The hierarchical hidden Markov model (HHMM) is employed to represent the models, in which each model is composed of a submodel of the target action and submodels which can represent any action, and they are connected appropriately. Several experimental results are shown.