Object tracking based on an online learning network with total error rate minimization

This paper presents a visual object tracking system which is tolerant to external imaging factors such as illumination, scale, rotation, occlusion and background changes. Specifically, an integration of an online version of total-error-rate minimization based projection network with an observation model of particle filter is proposed to effectively distinguish between the target object and the background. A re-weighting technique is proposed to stabilize the sampling of particle filter for stochastic propagation. For self-adaptation, an automatic updating scheme and extraction of training samples are proposed to adjust to system changes online. Our qualitative and quantitative experiments on 16 public video sequences show convincing performances in terms of tracking accuracy and computational efficiency over competing state-of-the-art algorithms.     

Video-based weathering gallery

We present a simple and effective method for constructing a gallery that consists of weathering effect elements called time-dependent appearance manifolds (TDAMs). Since TDAMs are computed from sample video clips showing dynamic weathering phenomena, they represent very smooth changes in the appearance of weathered pixels over time. Once a gallery with a variety of weathering effects is prepared, users can interactively choose and apply the predefined effects onto the surface of 3D graphic models and then finally assign the most appropriate one. This video-based weathering method can be implemented with very simple algorithms and it supports predictable, intuitive, and natural effects. Our system allows users to produce photorealistic augmented videos that include 3D graphic models weathered by our method. Moreover, users can easily enhance the realism of the augmented videos by manipulating rendering parameters such as the degree of weathering, texturing, lighting, and shadowing through user-friendly graphical user interfaces (GUIs).     

Nonparametric approach for uncertainty-based multidisciplinary design optimization considering limited data

Uncertainty-based multidisciplinary design optimization (UMDO) has been widely acknowledged as an advanced methodology to address competing objectives and reliable constraints of complex systems by coupling relationship of disciplines involved in the system. UMDO process consists of three parts. Two parts are to define the system with uncertainty and to formulate the design optimization problem. The third part is to quantitatively analyze the uncertainty of the system output considering the uncertainty propagation in the multidiscipline analysis. One of the major issues in the UMDO research is that the uncertainty propagation makes uncertainty analysis difficult in the complex system. The conventional methods are based on the parametric approach could possibly cause the error when the parametric approach has ill-estimated distribution because data is often insufficient or limited. Therefore, it is required to develop a nonparametric approach to directly use data. In this work, the nonparametric approach for uncertainty-based multidisciplinary design optimization considering limited data is proposed. To handle limited data, three processes are also adopted. To verify the performance of the proposed method, mathematical and engineering examples are illustrated.     

Quantification model of proper curvature for large‐sized curved TVs

We present a feasible modeling method to estimate the proper curvature for large‐sized curved TVs. We especially focused on two factors – preference and perceived distortion – to be considered to determine the proper curvature. The preference includes a number of advantages that people expect to fulfill the possibilities of a more immersive image and sense of realism from curved display. On the other hand, the perceived distortion means the side effects that people can notice uncomfortable feeling caused by the shape of a curved one. In order to find out how two factors would be different as a number of conditions like curvature, size, and viewing angle change, a series of subjective assessments were conducted. The evaluation results show that both the preference and perceived distortion vary with the conditions considerably. We performed the statistical analysis based on the results and proposed the quantification model of proper curvature, which has higher preference and less perceived distortion, for various‐sized curved TVs.     

Move blocking strategy applied to re-entrant manufacturing line scheduling

Model predictive control (MPC)-based approach to fab-wide scheduling has been suggested to solve constraint-aware production optimization and in-process inventory level control simultaneously at each scheduling instance. However, application of this approach to real fab suffers from computational difficulties brought by the need to solve a huge optimization problem on-line as real fab scheduling problems are characterized by long cycle times, multiple product types, hundreds of machines/processing steps and re-entrant product flows. This study explores the use of an offset-blocking strategy combined with a modified recursive least square (RLS) estimation in the fab-wide scheduler, in order to alleviate the difficulty. The strategy is tested on a modified version of published case study called Intel Mini-Fab (IMF) problem. Despite its simplicity, the blocking strategy showed excellent performance in the face of realistic demand changes and plant/model mismatch.     

Note: High speed optical profiler based on a phase-shifting technique using frequency-scanning lasers

We present a high speed optical profiler (HSOP) using frequency-scanning lasers for three-dimensional profile measurements of microscopic structures. To improve upon previous techniques for implementing the HSOP, we developed frequency-scanning lasers and a compact microscopic interferometer. The controller of the HSOP was also modified to generate proper phase-shifting steps. For measurements of step height specimens, the HSOP showed results comparable with a commercial optical profiler, even with much higher measurement speeds (up to 30 Hz). The typical repeatability of step height measurement was less than 1 nm. We also present measurements of microscopic structures to verify the HSOP's ability to perform high speed inline inspection for the semiconductor and flat-panel display industries.     

Performance Comparisons Between Thumb-Based and Finger-Based Input on a Small Touch-Screen Under Realistic Variability

The potential benefits of thumb-based touch interaction have not been fully exploited due to its usability problems and performance deterioration. Despite the well-known problems, mobile phone users often prefer thumb-based input method in their daily context of use. Without understanding input performance under realistic variability, design solutions may not address the problems adequately. This research aims to evaluate performance of one-handed thumb-based input compared to cradled finger-based input for the large number of users and varying task conditions. By investigating performance under a range of user- and task-variability, common patterns can be identified to help infer realistic performance in context of use. For this experiment, 259 participants were recruited balanced on gender and age. They performed user testing of moving an icon on a mobile touch-screen. Overall, the one-handed thumb input showed a 30% reduction in throughput compared to the cradled finger-based input, with significant reduction in speed and accuracy. Reduced throughput is attributed to inaccuracy rather than speed. In addition, the partial effects of touch position, dragging direction, and target size were investigated and quantified. In conclusion, performance could maintain constant throughput only for the finger-based input of limited task conditions, when realistic variability was introduced. Also, high variance of throughput for the thumb-based input led to poor conformity to Fitts’s law. The findings have implications for design of thumb-based touch interface to offset performance reduction and characterizing performance measure for thumb-based input method.     

An Integrated Model of the Toyota Production System with Total Quality Management and People Factors

The Toyota production system (TPS), or lean production, has been associated with many benefits for manufacturing firms that implement the system. However, to implement the TPS successfully, it is necessary to integrate the so‐called “hard side” of the system (that is, the technical aspects of material handling) with the “soft side” of the program (that is, the aspects associated with human factors). The present study makes a contribution to such a holistic view of the TPS by proposing an integrated model that consists of the technical aspects of the TPS, together with elements associated with total quality management (TQM), human resources management (HRM), and certain identified “people factors.” The study then uses structural equation modeling (SEM) to test this holistic model using data obtained from a questionnaire survey of 153 Taiwanese manufacturing firms. The empirical study confirms the appropriateness of the causal model, which is evaluated by several goodness‐of‐fit methods. The study confirms the proposition that the inclusion of the “people factors” together with TQM and HRM significantly improves the realization of benefits associated with the TPS practices of “just‐in‐time” (JIT) and “autonomation.” The conclusion is that such an integrated model provides a much more effective “lean system” and has the potential to produce significantly enhanced benefits for firms that implement it. © 2012 Wiley Periodicals, Inc.     

Peristaltic micropump system with piezoelectric actuators

This work presents a driving system for a peristaltic micropump that is based on piezoelectric actuation. The effects of the actuation sequence on pump performance are also considered. A valveless peristaltic micropump based on piezoelectric actuation is designed and fabricated using microelectromechanical system technology. The pump has three parts––silicon, Pyrex glass and commercially available bulk PZT (lead zirconate titanate) chips. The peristaltic micropump actuated by PZT chips comprises three chambers that are in series. The driving system consists of an ATmega 8535 microprocessor, a high voltage power supply, three differential amplifiers, a phase controller, an A/D converter, a 555 oscillator and an LCD module. It is supplied via a 110 Vrms 60-Hz AC line and is programmable. The system can produce step-function signals with voltages of up to 100 V_pp and frequencies ranging from 10 Hz to 1 kHz, as the inputs for the pump. Fluid pumping with air is successfully demonstrated. Additionally, 3-, 4- and 6-phase actuation sequences for the pump are designed and used to study the effects on pump performance, as revealed by the flow rate and the displacement of a pump diaphragm. The experimental results show that the flow rate and the displacement of the diaphragm actuated by the 4-phase sequence exceed those actuated by the 3- and 6-phase sequences. A flow rate of 17.6 μl min⁻¹ and a displacement of 2.91 μm (peak-to-peak) in 4-phase peristaltic motion are achieved at 100 Hz and 100 V_pp. The results demonstrate that the pump actuated in the 4-phase sequence is the most efficient. Consequently, the actuation sequences can affect the pump performance.