• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1287-1292
• /
• 2005

This paper presents an effective pattern classification model by designing an artificial neural network based pattern classifiers for face recognition. First, a RGB image inputted from a frame grabber is converted into a HSV image which is similar to the human beings' vision system. Then, the coarse facial region is extracted using the hue(H) and saturation(S) components except intensity(V) component which is sensitive to the environmental illumination. Next, the fine facial region extraction process is performed by matching with the edge and gray based templates. To make a light-invariant and qualified facial image, histogram equalization and intensity compensation processing using illumination plane are performed. The finally extracted and enhanced facial images are used for training the pattern classification models. The proposed H-ART2 model which has the hierarchical ART2 layers and F-LVQ model which is optimized by fuzzy membership make it possible to classify facial patterns by optimizing relations of clusters and searching clustered reference patterns effectively. Experimental results show that the proposed face recognition system is as good as the SVM model which is famous for face recognition field in recognition rate and even better in classification speed. Moreover high recognition rate could be acquired by combining the proposed neural classification models.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.477-484
• /
• 2001

This paper proposes a hybrid learning algorithm combined momentum and kernel-adatron for improving the performance of regression support vector machine. The momentum is utilized for high-speed convergence by restraining the oscillation in the process of converging to the optimal solution, and the kernel-adatron algorithm is also utilized for the capability by working in nonlinear feature spaces and the simple implementation. The proposed algorithm has been applied to the 1-dimension and 2-dimension nonlinear function regression problems. The simulation results show that the proposed algorithm has better the learning speed and performance of the regression, in comparison with those quadratic programming and kernel-adatron algorithm.

• Proceedings of the International Microelectronics And Packaging Society Conference
• /
• 2000.04a
• /
• pp.29-33
• /
• 2000

MEMS, Micro Electro-Mechanical Systems, present one of the greatest advanced packaging challenges of the next decade. Historically hybrid technology, generally thick film, provided sensors and actuators while integrated circuit technologies provided the microelectronics for interpretation and control of the sensor input and actuator output. Brought together in MEMS these technical fields create new opportunities for miniaturization and performance. Integrated circuit processing technologies combined with hybrid design systems yield innovative sensors and actuators for a variety of applications from single crystal silicon wafers. MEMS packages, far more simple in principle than today's electronic packages, provide only physical protection to the devices they house. However, they cannot interfere with the function of the devices and often must actually facilitate the performance of the device. For example, a pressure transducer may need to be open to atmospheric pressure on one side of the detector yet protected from contamination and blockage. Similarly, an optical device requires protection from contamination without optical attenuation or distortion being introduced. Despite impediments such as package standardization and complexity, MEMS markets expect to double by 2003 to more than ＄9 billion, largely driven by micro-fluidic applications in the medical arena. Like the semiconductor industry before it. MEMS present many diverse demands on the advanced packaging engineering community. With focused effort, particularly on standards and packaging process efficiency. MEMS may offer the greatest opportunity for technical advancement as well as profitability in advanced packaging in the first decade of the 21st century! This paper explores MEMS packaging opportunities and reviews specific technical challenges to be met.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.6
• /
• pp.416-424
• /
• 2014

This paper presents an 8-bit pipelined analog-to-digital converter. The supply voltage applied for comparators and other sub-blocks of the ADC were 0.7V and 0.5V, respectively. This low power ADC utilizes the capacitive charge pump technique combined with a source-follower and calibration to resolve the need for the opamp. The differential charge pump technique does not require any common mode feedback circuit. The entire structure of the ADC is based on fully dynamic circuits that enable the design of a very low power ADC. The ADC was designed to operate at 1MS/s in 90nm CMOS process, where simulated results using ADS2011 show the peak SNDR and SFDR of the ADC to be 47.8 dB (7.64 ENOB) and 59 dB respectively. The ADC consumes less than 1mW for all active dynamic and digital circuitries.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.10
• /
• pp.1044-1049
• /
• 2006

Multiple sequence alignment is a method to compare two or more DNA or protein sequences. Most of multiple sequence alignment tools rely on pairwise alignment and Smith-Waterman algorithm to generate an alignment hierarchy. Therefore, in the existing multiple alignment method as the number of sequences increases, the runtime increases exponentially. In order to remedy this problem, we adopted a parallel processing suffix tree algorithm that is able to search for common subsequences at one time without pairwise alignment. Also, the cross-matching subsequences triggering inexact-matching among the searched common subsequences might be produced. So, the cross-matching masking process was suggested in this paper. To identify the function of the clusters generated by suffix tree clustering, BLAST and CDD (Conserved Domain Database)search were combined with a clustering tool. Our clustering and annotating tool consists of constructing suffix tree, overlapping common subsequences, clustering gene sequences and annotating gene clusters by BLAST and CDD search. The system was successfully evaluated with 36 gene sequences in the pentose phosphate pathway, clustering 10 clusters, finding out representative common subsequences, and finally identifying functional domains by searching CDD database.

• Food Science and Biotechnology
• /
• v.16 no.2
• /
• pp.255-259
• /
• 2007

The effects of temperature heating-up rate and pressure building-up phase on the inactivation of Listeria innocua ATCC 33090 were evaluated in buffered peptone water. The number of L. innocua was reduced by 5.57 and 6.52 log CFU/mL during the nonisothermal treatment (the come-up time followed by isothermal process) and the isothermal treatment, respectively, at $60^{\circ}C$. When compared to the isothermal treatment (0.76$33.2^{\circ}C/min$ of temperature heating-rate. The effect of the combined high pressure and thermal processing on the inactivation of L. innocua increased with increasing pressure and temperature. At all temperature levels from 40 to $60^{\circ}C$ under 700 MPa, L. innocua was not detected by enrichment culture (>7 log reduction).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.10B
• /
• pp.1111-1116
• /
• 2009

In this paper, we discuss distortion-tolerant pattern recognition using computational integral imaging reconstruction. Three-dimensional object information is captured by the integral imaging pick-up process. The captured information is numerically reconstructed at arbitrary depth-levels by averaging the corresponding pixels. We apply Fisher linear discriminant analysis combined with principal component analysis to computationally reconstructed images for the distortion-tolerant recognition. Fisher linear discriminant analysis maximizes the discrimination capability between classes and principal component analysis reduces the dimensionality with the minimum mean squared errors between the original and the restored images. The presented methods provide the promising results for the classification of out-of-plane rotated objects.

• Korean Journal of Materials Research
• /
• v.20 no.10
• /
• pp.550-554
• /
• 2010

This work presents a fabrication procedure to make large-area, size-tunable, periodically different shape metal arrays using nanosphere lithography (NSL) combined with ashing and annealing. A polystyrene (PS, 580 ${\mu}m$) monolayer, which was used as a mask, was obtained with a mixed solution of PS in methanol by multi-step spin coating. The mask morphology was changed by oxygen RIE (Reactive Ion Etching) ashing and temperature processing by microwave heating. The Au or Pt deposition resulted in size tunable nano patterns with different morphologies such as hole and dots. These processes allow outstanding control of the size and morphology of the particles. Various sizes of hole patterns were obtained by reducing the size of the PS sphere through the ashing process, and by increasing the size of the PS sphere through annealing treatment, which resulted in tcontrolling the size of the metallic nanoparticles from 30 nm to 230 nm.

• Transactions of Materials Processing
• /
• v.20 no.5
• /
• pp.377-386
• /
• 2011

During stamping processes, the air trapped between sheet metal and the die cavity can be highly compressed and ultimately reduce the shape accuracy of formed panels. To prevent this problem, vent holes and passages are sometimes drilled into the based on expert experience and know-how. CAE can be also used for analyzing the air behavior in die cavity during stamping process, incorporating both elasto-plastic behavior of sheet metal and the fluid dynamic behavior of air. This study presents sheet metal forming simulation combined simultaneously with simulation of air behavior in the die cavity. There are three approaches in modeling of air behavior. One is a simple assumption of the bulk modulus having a constant pressure depending on volume change. The next is the use of the ideal gas law having uniform pressure and temperature in air domain. The third is FPM (Finite point method) having non-uniform pressure in air domain. This approach enables direct coupling of mechanical behavior of solid sheet metal and the fluid behavior of air in sheet metal forming simulation, and its result provides the first-hand idea for the location, size and number of the vent holes. In this study, commercial software, PAM-$STAMP^{TM}$ and PAM-$SAFE^{TM}$, were used.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.9
• /
• pp.878-886
• /
• 2012

This paper investigates the implementation and flight test of realtime vision guided autopilot system based on virtual instrumentation platform. A graphical design process via virtual instrumentation platform is fully used for the image processing, communication between systems, vehicle dynamics control, and vision coupled guidance algorithms. A significatnt ojective of the algorithm is to achieve an environment robust autopilot despite wind and an irregular image acquisition condition. For a robust vision guided path tracking and hovering performance, the flight path guidance logic is combined in a multi conditional basis with the position estimation algorithm coupled with the vehicle attitude dynamics. An onboard flight test equipped with the developed realtime vision guided autopilot system is done using the rotary UAV system with full attitude control capability. Outdoor flight test demonstrated that the designed vision guided autopilot system succeeded in UAV's hovering on top of ground target within about several meters under geenral windy environment.