• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.10
• /
• pp.3482-3497
• /
• 2021

Artificial intelligence has emerged as the core of the 4th industrial revolution, and large amounts of data processing, such as big data technology and rapid data analysis, are inevitable. The most fundamental and universal data interpretation technique is an analysis of information through regression, which is also the basis of machine learning. Ridge regression is a technique of regression that decreases sensitivity to unique or outlier information. The time-consuming calculation portion of the matrix computation, however, basically includes the introduction of an inverse matrix. As the size of the matrix expands, the matrix solution method becomes a major challenge. In this paper, a new algorithm is introduced to enhance the speed of ridge regression estimator calculation through series expansion and computation recycle without adopting an inverse matrix in the calculation process or other factorization methods. In addition, the performances of the proposed algorithm and the existing algorithm were compared according to the matrix size. Overall, excellent speed-up of the proposed algorithm with good accuracy was demonstrated.

• Journal of Information Processing Systems
• /
• v.15 no.3
• /
• pp.550-569
• /
• 2019

We propose approaches for improving Bloom filter in terms of false positive probability and membership query speed. To reduce the false positive probability, we propose special type of additional Bloom filters that are used to handle false positives caused by the original Bloom filter. Implementing the proposed approach for a routing table lookup, we show that our approach reduces the routing table lookup time by up to 28% compared to the original Bloom filter by handling most false positives within the fast memory. We also introduce an approach for improving the membership query speed. Taking the hash table-like approach while storing only values, the proposed approach shows much faster membership query speed than the original Bloom filter (e.g., 34 times faster with 10 subsets). Even compared to a hash table, our approach reduces the routing table lookup time by up to 58%.

• Current Optics and Photonics
• /
• v.2 no.6
• /
• pp.595-605
• /
• 2018

The two-photon process of microscopy provides good spatial resolution and optical sectioning ability when observing quasi-static endogenous fluorescent tissue within an in vivo animal model skin. In order to extend the use of such systems, we developed a two-photon laser scanning microscopy system capable of also capturing $512{\times}512$ pixel images at 90 frames per second. This was made possible by incorporating a 72 facet polygon mirror which was mounted on a 55 kRPM motor to enhance the fast-scan axis speed in the horizontal direction. Using the enhanced temporal resolution of our high-speed two-photon laser scanning microscope, we show that rapid processes, such as fluorescently labeled erythrocytes moving in mouse blood flow at up to 1.2 mm/s, can be achieved.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.56 no.3
• /
• pp.129-135
• /
• 2007

This paper deals with speed-sensorless control of DC servo motor using Neuro-Fuzzy Observer. DC servo motor has very low rotor inertia and excellent response characteristic and it is very useful to control torque and speed. It is easy to detect the voltage and current and resolver or encoder is used to measure a rotor speed. But it has a limit as a driving speed to detect speed precisely. So it is problem to improve the performance of the driving system. To solve this problem, it is studied to detect a speed of DC servo motor without sensor. In particular, study on the method to estimate the speed using the observer is performed a lot. In this paper, the gain of the observer is properly set up using the Neuro-Fuzzy control and Neuro-Fuzzy Observer that have a superior transient characteristic and is easy to implement compared the existing method is designed. It calculates the differentiation of the rotor current directly using the rotor current measured in the DC servo motor and estimates the speed of the rotor using the differentiation. Proposed speed sensorless control method is performed using the estimated speed. Also, it is proved feasibility of the proposed observer from the comparison tested a case with a speed sensor and a case without a speed sensor which used a highly efficient drive and 200[w] DC servo motor starting system.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.1
• /
• pp.10-15
• /
• 2015

This paper proposes the parallel implementation of a self-similarity based image SR (super-resolution) algorithm using OpenCL. The SR algorithm requires tremendous computations to search for a similar patch. This becomes a bottleneck for the real-time conversion from a FHD image to UHD. Therefore, it is imperative to accelerate the processing speed of SR algorithms. For parallelization, the SR process is divided into several kernels, and memory optimization is performed. In addition, two GPUs are used for further acceleration. The experimental results shows that a GPGPU implementation can speed up over 140 times compared to a single-core CPU. Furthermore, it was confirmed experimentally that utilizing two GPUs can speed up the execution time proportionally, up to 277 times.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.41 no.7
• /
• pp.532-538
• /
• 2013

This paper describes a vision-based only attitude estimation technique for the leader in the formation flight. The feature points in image obtained from the X-PLANE simulator are extracted by the SURF(Speed Up Robust Features) algorithm. We use POSIT(Pose from Orthography and Scaling with Iteration) algorithm to estimate attitude. Finally we verify that attitude estimation using vision only can yield small estimated error of $1.1{\sim}1.76^{\circ}$.

• Journal of Korean Society of Water and Wastewater
• /
• v.26 no.6
• /
• pp.851-860
• /
• 2012

In this study, to evaluate the abatement of runoff pollution loads by the road sweeping(cleaning), various investigations are implemented at the sample area of the highway. As the results of evaluating the removal efficiency of pollutants along road cleaning, TSS showed about 78 % of the removal efficiency and COD showed 49 % of removal efficiency through the operation of cleaning vehicle of vacuum suction method. In case of TN and TP, they showed the relatively-lower removal efficiency by 30~35 %. TSS removal efficiency along the number of cleaning appeared about 60 % in case of one time of cleaning and the additional removal effect did not appear though the number of cleaning increased to two times. With running speed of cleaning vehicle, TSS removal ratio is lessened from 60 % to 20 % when cleaning vehicle speed up to 20 km/hr from 6 km/hr. It seems that the reasons why the removal efficiencies are inversely proportional to its speed are related to the lower vacuum efficiencies and the disturbed particles on the road. In the pollutant build-up analysis, it is showed that it takes more time to the critical pollutant build-up in the shoulder than the center of the road. It is also showed that the proper cleaning cycle is recommended as 4~6 dry weather days without rainfall events.

• International Journal of Automotive Technology
• /
• v.7 no.5
• /
• pp.527-534
• /
• 2006

Modern passenger cars with diesel engines are equipped with DOC(diesel oxidation catalyst) for the purpose of reducing HC and CO in the exhaust stream. Cold start exhaust emissions pose troubles here as on gasoline engine vehicles. As a result, some of the diesel passenger cars roll off todays the assembly lines with WCC(warm-up catalytic converter). Oxidation characteristics of the particulates in WCC is analyzed in this study by EEPS(engine exhaust particulate size spectrometer). The maximum number of PM is found to come out of WCC in sizes near 10nm when an HSDI diesel engine is operated under the conditions of high speed and medium to heavy load. When the temperature of the WCC exceeds $300^{\circ}C$, the number of PM smaller than 30 nm in diameter sharply increases upon passing through the WCC. Total mass of emitted PM gets reduced downstream of the WCC under low speed and light load conditions due to adsorption of PM onto the catalyst. Under conditions of high speed and medium to heavy load, the relatively large PM shrink or break into fine particles during oxidation process within the WCC, which results in more mass fraction of fine particles downstream of the WCC.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.3
• /
• pp.74-80
• /
• 2004

If the withdrawal force of strand driven roll which is controlled with predetermined speed isn't distributed regularly in continuous casting process, mold level is changed cyclically and slab quality is not good. In this paper, both casting speed control and load sharing control algorithm of strand driven roll according to ratio which is set up by operator at same time is proposed. The proposed algorithm is to share the withdrawal force by following torque current of each driven roll as average torque current by changing speed reference of each driven roll motor. The load sharing control with anti-wind up for improvement transient state is adopted and the proposed algorithm is implemented in POSCO pilot caster.

• Fashion & Textile Research Journal
• /
• v.1 no.2
• /
• pp.166-172
• /
• 1999

An experimental investigation has done for studying the changes of physical properties of the air-jet textured polyester yarn. Through the experiments of polyester 85/72 drawn yarn and 50/36 partially oriented yarn doubling' the following results are obtained. The tenacity of textured yarn decreased as the number of tension rings increased when the nozzle diameter was 1.2 mm. In the case when the tension ring was not used during the texturing process, the tenacity of textured yarn decreased as the nozzle diameter increased. But this tendency changed according to the number of tension rings used, the diameter of the nozzle, and yarn take-up speed. Breaking elongation decreased as the number of tension rings increased from one to four and the take-up speed increased at diameter 1.2 mm of the nozzle. The higher the diameter of the the nozzle the higher the breaking elongation until 1.4 mm. The number of interlacing went down as the diameter of the nozzle increased and the take-up speed of the yarn increased. The higher the number of tension rings, the lower the number of interlacing.