• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.9
• /
• pp.235-244
• /
• 2016

Automotive symbols are more widely needed for new, convenient driving devices in automobiles. Good automotive symbols should be detectable, identifiable at first glance, easily learned, recognizable, and produce quick responses after practice. In this paper, a methodology for developing and evaluating automotive symbols is suggested. It includes multiple tests, such as comprehension, perceptual quality, appropriateness, and integrated evaluation. 28 symbols were tested and evaluated by the suggested methodology for convenient driving systems, such as a lane departure warning system (LDWS), cruise control (CCS), and a collision warning system (CWS). Most of the KS R ISO 2575 symbols had higher scores of comprehension, perceptual quality, and appropriateness, but the sunroof and camera symbols had lower scores. Standard symbols with several new functions should be developed. This methodology could be useful for developing and evaluating automotive symbols.

• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we propose an autonomous driving system using an end-to-end model to improve lane departure and misrecognition of traffic lights in a vision sensor-based system. End-to-end learning can be extended to a variety of environmental conditions. Driving data is collected using a model car based on a vision sensor. Using the collected data, it is composed of existing data and data with outlayers removed. A class was formed with camera image data as input data and speed and steering data as output data, and data learning was performed using an end-to-end model. The reliability of the trained model was verified. Apply the learned end-to-end model to the model car to predict the steering angle with image data. As a result of the learning of the model car, it can be seen that the model with the outlayer removed is improved than the existing model.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.40 no.2
• /
• pp.167-174
• /
• 1995

This experiment was carried out to investigate the major characters affecting the popping volume of popcorn. Tuygimok 1 (Kp1 ${\times}$ Kp2) and 8 popcorn hybrids' agronomic characters were tested to evaluate a certain extent how much they affect on the popping volume. Moisture con-tent was considered as the most important factor, but failed to evaluate the optimum moisture con-tent level in this experiment moisture range (12.2-14.4%) because popping volume increased as moisture content of kernels increased. The maximum popping volume was obtained at 55-60kg of kernel hardness, 80-90,um of pericarp thickness and 45-50% of S/H (Soft/Hard starch). But the Em/En(Embryo/Endosperm) ratio was negatively associated with the popping volume. Therefore the minimum popping volume was observed at the 10-11 % of Em/En ratio. Moisture content, hardness, pericarp thickness, Em/En and S/H ratio were selected as the appropriate variables for the maxi-mum popping volume using the stepwise forward regression method and the expecting popping volume was estimated by the multiple linear regression formular. The mean popping volume of ninepopcorn hybrids was about 29.2cm3/g.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.3
• /
• pp.80-92
• /
• 2008

Many predominant video coding tools in terms of coding efficiency were adopted in the latest video coding standard, H.264/AVC. Regardless of development of these predominant video coding tools such as the variable block-size motion estimation/compensation, intra prediction based on various directions, and so on, the discrete cosine transform has been continuously used starting from the early video coding standards. Generally, the correlation coefficient of the residual signal is usually less than 0.5 when this residual signal is actually encoded. In this interval of correlation coefficient, the discrete cosine transform does not show the optimal coding gain, and the discrete sine transform which is a sub-optimal transform when the correlation coefficient is in the interval from -0.5 to 0.5 can be used in conjunction with the discrete cosine transform in the video coding. In this paper, an alternative transform that alternatively uses the discrete sine transform and integer cosine transform in H.264/AVC by using rate-distortion optimization is proposed. The proposed method achieves a BD-PSNR gain of up to 0.71 dB compared to H.264/AVC JM 10.2 at relatively high bitrates.

• Journal of Technology Innovation
• /
• v.26 no.3
• /
• pp.37-68
• /
• 2018

Since the 20th century, automobiles, which are the most common means of transportation, have been evolving as the use of electronic control devices and automotive semiconductors increases dramatically. Automotive semiconductors are a key component in automotive electronic control devices and are used to provide stability, efficiency of fuel use, and stability of operation to consumers. For example, automotive semiconductors include engines control, technologies for managing electric motors, transmission control units, hybrid vehicle control, start/stop systems, electronic motor control, automotive radar and LIDAR, smart head lamps, head-up displays, lane keeping systems. As such, semiconductors are being applied to almost all electronic control devices that make up an automobile, and they are creating more effects than simply combining mechanical devices. Since automotive semiconductors have a high data rate basically, a microprocessor unit is being used instead of a micro control unit. For example, semiconductors based on ARM processors are being used in telematics, audio/video multi-medias and navigation. Automotive semiconductors require characteristics such as high reliability, durability and long-term supply, considering the period of use of the automobile for more than 10 years. The reliability of automotive semiconductors is directly linked to the safety of automobiles. The semiconductor industry uses JEDEC and AEC standards to evaluate the reliability of automotive semiconductors. In addition, the life expectancy of the product is estimated at the early stage of development and at the early stage of mass production by using the reliability test method and results that are presented as standard in the automobile industry. However, there are limitations in predicting the failure rate caused by various parameters such as customer's various conditions of use and usage time. To overcome these limitations, much research has been done in academia and industry. Among them, researches using data mining techniques have been carried out in many semiconductor fields, but application and research on automotive semiconductors have not yet been studied. In this regard, this study investigates the relationship between data generated during semiconductor assembly and package test process by using data mining technique, and uses data mining technique suitable for predicting potential failure rate using customer bad data.

• Journal of Korean Society of Transportation
• /
• v.14 no.2
• /
• pp.59-88
• /
• 1996

This study has determined optimal type, and location of loop detector to measure accurately traffic condition influenced by traffic variation with real time. Optimal type of loop detector for through vehicle at stop bar was determined by confidences of occupancy period, and nonoccupancy period, and so appropriate detector type for application to real time traffic control system has been decided on special loop detector.

• 전자공학회논문지 IE
• /
• v.44 no.1
• /
• pp.27-32
• /
• 2007

This study was to suggest the predictive model of propagation, considering the effect by the multipath waves produced by the sending and receiving vehicles' left/right reflectors and the adjacent vehicles when the communication between the vehicles on the one-way two-lanes road in the urban city with a lot of traffic jams. Then, the radius of curved road was 600[m], the length of curved roads $52.4\sim471.2[m]$, and the bridge's pier of road was $5o\sim45o$. Also, it was simulated by changing the receiving vehicle located on the curved road's gap from minimum 3.3[m] to maximum 29.5[m], corresponding to the change of distance of the bridge's pier of road and curved road. As a result of this research above, in case of $5o\sim15o$ bridge's pier of road, it was within l[dB] regardless of the receiving vehicle's position on the curved road in case of propagation path loss. In case of $15o\sim45o$, it was approximately $1\sim8[dB]$ as the bridge's pier of road is changed. And, in case of propagation path, it found out that it was changed to $0.4\sim120[m]$ according to the change of bridge's pier of road. Then, the delay time of propagation was 400[nsec] as it produced 120[m] in the difference of propagation path.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.22 no.6
• /
• pp.105-113
• /
• 2018

The proposed model to predict the bridge load carrying capacity uses the impact response spectrum. The spectrum is based on Euler-Bernoulli beam and the center of the bridge width for the moving load location. Therefore, it is necessary to investigate the eccentric moving load effects on the impact factor and response factor. For this, this study considers 10 m width and two-lane simply supported slab bridges and performs the moving load analysis to investigate the variations of peak impact factor and corresponding response factor. The numerical results show that the eccentric load increases both the static and dynamic displacements, but the impact factor is decreased since the incremental amount of static displacement is bigger than that of dynamic displacement. However, the difference of the impact factors between the center and eccentric loadings is small showing less than 0.5%p. In the response factor, the eccentric loading increases both the static and dynamic response factors, compared to the center loading. The difference of the response factor is only 0.18%p. It shows that the eccentric loading has very small effects on the response factor, thus the impact factor response spectrum which is generated based on the center moving load can be used to determine the response factor.

• Korean Journal of Occupational Therapy
• /
• v.26 no.4
• /
• pp.127-143
• /
• 2018

Objective : The purpose of this study was to compare with the effects of driving simulator and Dynavision training after stroke through the test of cognitive ability and driving performance. Methods : Twenty-one stroke patients were randomly classified to the driving simulator training group (N=11) and Dynavision training group (N=10), and were carried out respectively training for 15 times. The driving performances was measured by the driving simulator test, and cognitive-perceptive abilities was measured by the DriveABLE Cognitive Assessment Tool, Trail Making Test-A, Trail Making Test-B and Mini Mental State Examination-K. Results : The driving simulator training group showed significant changes in all cognitive tests and most of driving performances. The Dynavision training group also showed significant changes in all cognitive tests except for Trail Making Test-A and some driving performances. The significant differences on both groups were found regarding the estimated degree of results on the on-road evaluation, the number of off road accidents and collisions. In addition, the causal influence of the two training methods on these variables was analyzed to be more than 20%. Conclusion : The driving simulator and Dynavision training were found to be effective intervention in the driving rehabilitation after stroke. In particular, it was confirmed that the driving simulator is an effective training to improve overall driving ability of stroke patients. In addition, the difference in training effect between the two training methods was found to be more than 20%.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.419-432
• /
• 2019

Most of deep learning model training was proceeded by supervised learning, which is to train labeling data composed by inputs and corresponding outputs. Labeling data was directly generated manually, so labeling accuracy of data is relatively high. However, it requires heavy efforts in securing data because of cost and time. Additionally, the main goal of supervised learning is to improve detection performance for 'True Positive' data but not to reduce occurrence of 'False Positive' data. In this paper, the occurrence of unpredictable 'False Positive' appears by trained modes with labeling data and 'True Positive' data in monitoring of deep learning-based CCTV accident detection system, which is under operation at a tunnel monitoring center. Those types of 'False Positive' to 'fire' or 'person' objects were frequently taking place for lights of working vehicle, reflecting sunlight at tunnel entrance, long black feature which occurs to the part of lane or car, etc. To solve this problem, a deep learning model was developed by simultaneously training the 'False Positive' data generated in the field and the labeling data. As a result, in comparison with the model that was trained only by the existing labeling data, the re-inference performance with respect to the labeling data was improved. In addition, re-inference of the 'False Positive' data shows that the number of 'False Positive' for the persons were more reduced in case of training model including many 'False Positive' data. By training of the 'False Positive' data, the capability of field application of the deep learning model was improved automatically.