• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.3
• /
• pp.59-72
• /
• 2017

As the part of study which handles the measure to use the individual vehicle information of taxi GPS data on signal controls in order to overcome the limitation of Loop detector-based collecting methods of real-time signal control system, this paper conducted series of evaluations and improvements on link travel time, queue vehicle time estimates and traffic condition decision algorithm from the research introduced in 2016. considering the control group and the other, the link travel time has enhanced the travel time and the length of queue vehicle has enhanced the estimated model taking account of the traffic situation. It is analyzed that the accuracy of the average link travel time and the length of queue vehicle are respectably both approximately 95 % and 85%. The traffic condition decision algorithm reflected the improved travel speed and vehicle length. Smoothing was performed to determine the trend of the traffic situation and reduce the fluctuation of the data, and the algorithms have refined so as to reflect the pass period on overflow judgment criterion.

• Speech Sciences
• /
• v.5 no.1
• /
• pp.7-21
• /
• 1999

This paper proposes an improved training procedure in speech recognition based on the continuous density of the Hidden Markov Model (CDHMM). Of the three parameters (initial state distribution probability, state transition probability, output probability density function (p.d.f.) of state) governing the CDHMM model, we focus on the third parameter and propose an efficient algorithm that determines the p.d.f. of each state. It is known that the resulting CDHMM model converges to a local maximum point of parameter estimation via the iterative Expectation Maximization procedure. Specifically, we propose two independent algorithms that can be embedded in the segmental K -means training procedure by replacing relevant key steps; the adaptation of the number of mixture Gaussian p.d.f. and the initialization using the CDHMM parameters previously estimated. The proposed adaptation algorithm searches for the optimal number of mixture Gaussian humps to ensure that the p.d.f. is consistently re-estimated, enabling the model to converge toward the global maximum point. By applying an appropriate threshold value, which measures the amount of collective changes of weighted variances, the optimized number of mixture Gaussian branch is determined. The initialization algorithm essentially exploits the CDHMM parameters previously estimated and uses them as the basis for the current initial segmentation subroutine. It captures the trend of previous training history whereas the uniform segmentation decimates it. The recognition performance of the proposed adaptation procedures along with the suggested initialization is verified to be always better than that of existing training procedure using fixed number of mixture Gaussian p.d.f.

• Journal of The Korean Association of Information Education
• /
• v.25 no.5
• /
• pp.813-824
• /
• 2021

To improve educational performance by analyzing the learning trends of adult learners of Open High Schools, various algorithm models using artificial intelligence were designed and performance was evaluated by applying them to real data. We analyzed Log data of 115 adult learners in the cyber education system of Open High Schools. Most adult learners of Open High Schools learned more than recommended learning time, but at the end of the semester, the actual learning time was significantly reduced compared to the recommended learning time. In the second half of learning, the participation rate of VODs, formation assessments, and learning activities also decreased. Therefore, in order to improve educational performance, learning time should be supported to continue in the second half. In the latter half, we developed an artificial intelligence algorithm models using Tensorflow to predict learning time by data they started taking the course. As a result, when using CNN(Convolutional Neural Network) model to predict single or multiple outputs, the mean-absolute-error is lowest compared to other models.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1551-1557
• /
• 2019

In modern society, a variety of video media have been widely spread in line with the fourth industrial revolution and the development of IoT technology; in accordance with this trend, numerous researches have been performed to remove noise generated in image and data communications. However, the conventional Additive White Gaussian Noise (AWGN) cancellation techniques are likely to induce a blurring phenomenon in the noise removal process, thus impairing the information of the image. In this study, we propose an algorithm for minimizing the loss of image information in the removal process of AWGN. The proposed algorithm can apply weights according to the characteristics of noise by predicting AWGN in the image, where the output is calculated based on adding and subtracting the outputs of the high pass filter and the low pass filter. Compared to the existing method, the noise reduction using the proposed algorithm exhibited less blurring issues and better noise reduction properties in the AWGN removal process.

• Journal of Platform Technology
• /
• v.8 no.2
• /
• pp.10-21
• /
• 2020

The use of WiFi is gradually increasing through the spread of mobile terminals and the development of data communication. Mobile Internet usage has been steadily increasing from the 2000s to the present. Almost all households in Korea have smart devices, and 90% of the population uses mobile Internet. Due to this trend, the government is currently constructing public WiFi zones in dense urban areas as a way to reduce communication costs. The WiFi usage in the public WiFi zone is increasing every year. Therefore, in this paper, we propose a method for using such public WiFi efficiently. A mobile terminal collects WiFi information and constructs a WiFi zone in a map using a concave hull algorithm. In the map, the mobile terminal provides a route through many public WiFi areas. As a result, the WiFi usage of the mobile terminal is increased through more WiFi regions.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.8
• /
• pp.1117-1123
• /
• 2021

Object recognition is to identify the location and class of a specific object by analyzing the given image when a specific image is input. One of the fields in which object recognition technology is actively applied in recent years is autonomous vehicles, and this paper describes the trend of image-based object recognition artificial intelligence technology in autonomous vehicles. The image-based object detection algorithm has recently been narrowed down to two methods (a single-step detection method and a two-step detection method), and we will analyze and organize them around this. The advantages and disadvantages of the two detection methods are analyzed and presented, and the YOLO/SSD algorithm belonging to the single-step detection method and the R-CNN/Faster R-CNN algorithm belonging to the two-step detection method are analyzed and described. This will allow the algorithms suitable for each object recognition application required for autonomous driving to be selectively selected and R&D.

• Progress in Medical Physics
• /
• v.27 no.4
• /
• pp.180-188
• /
• 2016

Acuros XB advanced dose calculation algorithm (AXB, Varian Medical Systems, Palo Alto, CA) has been released recently and provided the advantages of speed and accuracy for dose calculation. For clinical use, it is important to investigate the dosimetric performance of AXB compared to the calculation algorithm of the previous version, Anisotropic Analytical Algorithm (AAA, Varian Medical Systems, Palo Alto, CA). Ten volumetric modulated arc therapy (VMAT) plans for each of the following cases were included: head and neck (H&N), prostate, spine, and lung. The spine and lung cases were treated with stereotactic body radiation therapy (SBRT) technique. For all cases, the dose distributions were calculated using AAA and two dose reporting modes in AXB (dose-to-water, $AXB_w$, and dose-to-medium, $AXB_m$) with same plan parameters. For dosimetric evaluation, the dose-volumetric parameters were calculated for each planning target volume (PTV) and interested normal organs. The differences between AAA and AXB were statistically calculated with paired t-test. As a general trend, $AXB_w$ and $AXB_m$ showed dose underestimation as compared with AAA, which did not exceed within -3.5% and -4.5%, respectively. The maximum dose of PTV calculated by $AXB_w$ and $AXB_m$ was tended to be overestimated with the relative dose difference ranged from 1.6% to 4.6% for all cases. The absolute mean values of the relative dose differences were $1.1{\pm}1.2%$ and $2.0{\pm}1.2%$ when comparing between AAA and $AXB_w$, and AAA and $AXB_m$, respectively. For almost dose-volumetric parameters of PTV, the relative dose differences are statistically significant while there are no statistical significance for normal tissues. Both $AXB_w$ and $AXB_m$ was tended to underestimate dose for PTV and normal tissues compared to AAA. For analyzing two dose reporting modes in AXB, the dose distribution calculated by $AXB_w$ was similar to those of AAA when comparing the dose distributions between AAA and $AXB_m$.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2000.04b
• /
• pp.159-167
• /
• 2000

In optimization problems, computationally intensive or expensive simulations hinder the use of standard optimization techniques because the computational expense is too heavy to implement them at each iteration of the optimization algorithm. Therefore, those expensive simulations are often replaced with approximation models which can be evaluated nearly free. However, because of the limited accuracy of the approximation models, it is practically impossible to find an exact optimal point of the original problem. Significant efforts have been made to overcome this problem. The approximation models are sequentially updated during the iterative optimization process such that interesting design points are included. The interesting points have a strong influence on making the approximation model capture an overall trend of the original function or improving the accuracy of the approximation in the vicinity of a minimizer. They are successively determined at each iteration by utilizing the predictive ability of the approximation model. This paper will focuses on those approaches and introduces various approximation methods.

• Proceedings of the IEEK Conference
• /
• 2000.11d
• /
• pp.123-126
• /
• 2000

In this paper, the portable game machine called W"alking Beat" is designed and implemented not only to propose the new possibilities for the peripheral equipment market of portable acoustic machine but also to overcome the limitation of the acoustic simulation game machine such as the existing Beat Mania. The old game machine can be used only in a particular place, where it is installed. However, in order to get over the constraint on this space problem "Walking Beat Game Machine" is designed to facilitate the portability. In addition, the frequency analysis method using FFT algorithm is employed by regarding the music data for the portable digital acoustic machine as source so that the limitation that the existing game machine depends heavily on the previously registered game contents can be overcome. By making it possible to play games for all the music and putting an emphasis on multimedia trend only to listen to the music, it is speculated that it can contribute to the development of the new culture in the near future.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.411-414
• /
• 2005

The most recent trend on digital hearing aid is to increase the ratio of signal to noise by directivity or to develop noise reduction algorithm inside DSP IC chip. This paper designed, fabricated and tested a digital hearing aid directivity testing device in which a micro-mouse-like the stepping motor with a speaker rotates around an examinant. Both ears of the examinant were fixed with ITE hearing aids in order to response to receiving sound. The diameter of the directivity testing device was 2 [m] and the micro-mouse was precisely controlled by PICBASIC micro processor.