• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2001.05a
• /
• pp.265-268
• /
• 2001

In this paper, we designed one-dimensional VLSI array with high speed processing in Fractal image compression. fractal image compression algorithm partitions the original image into domain blocks and range blocks then compresses data using the self similarity of blocks. The image is partitioned into domain block with 50% overlapping. Domain block is reduced by averaging the original image to size of range block. VLSI array is trying to search the best matching between a range block and a large amount of domain blocks. Adjacent domain blocks are overlapped, so we can improve of each block's processing speed using the reuse of the overlapped data. In our experiment, proposed VLSI array has about 25% speed up by adding the least register, MUX, and DEMUX to the PE.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• Journal of Microbiology and Biotechnology
• /
• v.13 no.5
• /
• pp.823-827
• /
• 2003

Streptomyces venezuelae ATCC 15439 is notable in its ability to produce two distinct groups of macrolactones. It has been reported that the generation of two macrolactone structures results from alternative expression of pikromycin (Pik) polyketide synthase (PKS). It was previously reported that the hybrid pikromycin-tylosin PKS can also produce two different macrolactones but its mechanistic basis remains unclear. In order to address this question, a series of site-directed mutagenesis of tentative alternative ribosome binding site and translation start codons in tylGV were performed. The results suggest that macrolactone ring size is not determined by the alternative expression of TylGV but through other mechanism(s) involving direct interaction between the PikAIII and TE domain or skipping of the final chain elongation step. This provides new insight into the mechanism of macrolactone ring size determination in hybrid PKS as well as an opportunity to develop novel termination activities for combinatorial biosynthesis.

• Journal of The Korean Association For Science Education
• /
• v.16 no.2
• /
• pp.134-145
• /
• 1996

The purpose of this study was to investigate effects of mental capacity and mental demand on problem solving. Two kinds of tests were used for this study. One was FIT 752 test which required general knowledge, the other was Ohm's law test which required domain specific knowledge. The items of each test had the same logical structure and content knowledge but had different sizes of mental demand. The results of the study were summarized as follows: As mental demand of the items increased, the success rates decreased. The analysis of the hierarchical relation among items was that items with large mental demand were higher than those with small demand or at least the same level. According to the results, mental of an item was a significant factor affecting solving the problem. Effect of mental capacity on problem solving was different according to the kind of required knowledge to solve. Mental capacity was a significant factor affecting solving the FIT 752 task which required general knowledge. On the contary, solving the task which required domain specific knowledge, the results were different depending on subjects' chunk size. The results of problem solving of the groups which had small chunk size were that mental capacity was appeared a significant factor. However, results of problem solving of groups which had large chunk size were that mental capacity was not.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2018.06a
• /
• pp.32.1-32.1
• /
• 2018

Novel soft magnetic materials can be achieved by altering material properties such as morphology, composition, crystallinity, and grain size of soft magnetic alloys. Especially, magnetic properties (i.e., saturation magnetization, coarcivity) of soft magnetics are significantly affected by grain boundaries which act as a control of magnetic domain wall movement. Thus, we herein develop a two-step electroless plating method to control morphology and grain size of FeCo films for excellent magnetic properties. Accordingly, the chemical composition to control the degree of polarization of FeCo alloys was altered by electroless deposition parameters; for example, electrolyte concentration and temperature. The grain size and crystallinity of FeCo alloys was dramatically affected by the reaction temperature because the grain growth mechanism dominantly occurs at $90^{\circ}C$ where as the neucleation only happens at $50^{\circ}C$. By simply controlling the temperature, the micron-sized FeCo grains embedded FeCo film was synthesized where the large grains allow high magnetization originated from larger magnetic domain with low corecivity and the nano-sized grains allow excellent soft magnetic properties due to the magnetic correlation length.

• Korean Journal of Cognitive Science
• /
• v.26 no.4
• /
• pp.377-392
• /
• 2015

Korean semantic role labeling (SRL) is usually performed by a machine learning and requires a lot of corpus. However, the Korean PropBank used in Korean SRL system is less than PropBank. It leads to a low performance. Therefore, we expand the annotated corpus and verb frames for Korean SRL system to expand the Korean PropBank corpus. Most of the SRL system have a domain-dependent performance so, the performance may decrease if domain was changed. In this paper, we use the domain adaptation technique to reduce decreasing performance with the existing corpus and the small size of new domain corpus. We apply the domain adaptation technique to Structural SVM and Deep Neural Network. The experimental result show the effectiveness of the domain adaptation technique.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.11C
• /
• pp.917-924
• /
• 2008

Motion estimation and compensation in the spatial domain check the searching area of specified size in the previous frame and search block to minimize the difference with current block. When we check the searching area, it consumes the most encoding times due to increasing the complexity. We can solve this fault by means of motion estimation using shifting matrix in the transform domain instead of the spatial domain. We derive so the existed shifting matrix to a new recursion equation that we decrease more computations. We modify simply vertical shifting matrix and horizontal shifting matrix in the transform domain for motion estimation of half-pixel accuracy. So, we solve increasing computation due to bilinear interpolation in the spatial domain. Simulation results prove that motion estimation by the proposed algorithm in DCT-based transform domain provides higher PSNR using fewer bits than results in the spatial domain.

• The Journal of the Acoustical Society of Korea
• /
• v.30 no.6
• /
• pp.303-307
• /
• 2011

This paper describes an audio watermarking using Quantization Index Modulation (QIM) on significant peaks in frequency domain. The audio signal is broken up into L samples length frames with non-overlapping and rectangular window. The zero-crossing rate of each frame is calculated for decision whether it is proper to be watermarked or not. If the frame is legitimate, frequency magnitude response is computed by discrete Fourier transform. For the QIM, we set the quantization step size based on maximum value of frequency magnitude response and select n significant peaks with w samples around them in frequency domain, totally $n{\times}(w+1)$ samples. Finally, watermark embedding is performed. Decoder extract watermarks based on Euclidean distance, that is a blind detection. The proposed method is robust against many attacks of watermark benchmark.

• Journal of Educational Research in Mathematics
• /
• v.25 no.1
• /
• pp.113-137
• /
• 2015

The objective of the present study is to systematically examine the effects of on the cognitive and affective domains of elementary, middle, and high school students by conducting a meta-analysis. To this end, this study selected 31 research papers that had analyzed the effects of applying, and performed a meta-analysis of the findings presented in each research paper. The results obtained from the meta-analysis are presented as follows. First, both the collaborative learning program and the peer tutoring program for underachieving students in math manifested an above average size of effect in the cognitive domain. In particular, the effect was the greatest at the elementary school level, and out of the two programs, peer tutoring was identified to have a sizable effect. Second, both programs displayed an above average size of effect in the affective domain, and peer tutoring was identified to have a higher effect than collaborative learning. In addition, when the programs were compared based on school levels, the size of effect was highest at the elementary school level followed by middle school and high school, in that order. When compared based on the criteria of the affective domain, self-efficacy in math, learners' attitude toward math, and learners' interest in math were identified to. Finally, this study presented suggestions for teaching underachieving students in math and conducting follow-up studies based on the analysis results.

• Proceedings of the IEEK Conference
• /
• 2004.06a
• /
• pp.265-268
• /
• 2004

In this paper. we propose an adaptive step-size algorithm for the adaptive interference canceller (AIC) in the space-time trellis coded DS-CDMA system. In the AIC, the performance of the blind LMS algorithms that updates the tap-weight vector of the AIC is heavily dependent on the choice of step-size. To improve the performance of the fixed step-size AIC (FS-AIC), the regular adaptive step-size algorithm is extended in complex domain and applied to the joint AIC and ML decoder scheme. Simulation results show that the joint adaptive step-size AIC (AS-AIC) and ML decoder scheme using the proposed algorithm has boner performance than not only the conventional ML decoder but also the joint FS-AIC and ML decoder scheme without much increase of the decoding delay and complexity.