• Journal of KIISE
• /
• v.44 no.1
• /
• pp.44-50
• /
• 2017

Pedestrian detection, an important component of autonomous driving and driving assistant system, has been extensively studied for many years. In particular, image based pedestrian detection methods such as Hierarchical classifier or HOG and, deep models such as ConvNet are well studied. The evaluation score has increased by the various methods. However, pedestrian detection requires high sensitivity to errors, since small error can lead to life or death problems. Consequently, further reduction in pedestrian detection error rate of autonomous systems is required. We proposed a new method to detect pedestrians and reduce the error rate by using the Faster R-CNN with new developed pedestrian training data sets. Finally, we compared the proposed method with the previous models, in order to show the improvement of our method.

• The Journal of the Korea Contents Association
• /
• v.14 no.10
• /
• pp.527-538
• /
• 2014

The demand prediction is a critical issue for the film industry. As the social media, such as Twitter and Facebook, gains momentum of late, considerable efforts are being dedicated to prediction and analysis of hit movies based on unstructured text data. For prediction of trends found in commercially successful films, the correlations between the amount of data and hit movies may be analyzed by estimating the data variation by period while opinion mining that assigns sentiment polarity score to data may be employed. However, it is not possible to understand why the audience chooses a certain movie or which attribute of a movie is preferred by using such a quantitative approach. This has limited the efforts to identify factors driving a movie's commercial success. In this regard, this study aims to investigate a movie's attributes that reflect the interests of the audience. This would be done by extracting topic keywords that represent the contents of Twits through frequency measurement based on the collected Twitter data while analyzing responses displayed by the audience. The objective is to propose factors driving a movie's commercial success.

• Information Systems Review
• /
• v.6 no.2
• /
• pp.209-225
• /
• 2004

This study shows the field of construction which was less developed in the aspect of information system in comparison to other areas needs to introduce construction information system and suggests it can make more value linked with company performance through the system. For the first step, it emphasizes the integrated enterprise approach not optimization of each sector in a company. This paper extracts the current issues of construction information from the core process of construction area and implies a driving plan and the future development plan for construction information. Antecedent studies for this drew the information structure model and the construction information-driving model of a construction company. Hence, case studies with management companies that implemented the construction information system successfully tested validity. This research brings the cases of PMS, the integrated finance system, and CRM system as successful ones.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.4
• /
• pp.95-105
• /
• 2021

Road surface damage detection is essential for a comfortable driving environment and the prevention of safety accidents. Road management institutes are using automated technology-based inspection equipment and systems. As one of these automation technologies, a sensor to detect road surface damage plays an important role. For this purpose, several studies on sensors using deep learning have been conducted in recent years. Road images and label images are needed to develop such deep learning algorithms. On the other hand, considerable time and labor will be needed to secure label images. In this paper, the adversarial learning method, one of the semi-supervised learning techniques, was proposed to solve this problem. For its implementation, a lightweight deep neural network model was trained using 5,327 road images and 1,327 label images. After experimenting with 400 road images, a model with a mean intersection over a union of 80.54% and an F1 score of 77.85% was developed. Through this, a technology that can improve recognition performance by adding only road images was developed to learning without label images and is expected to be used as a technology for road surface management in the future.

• Journal of The Korean Society of Integrative Medicine
• /
• v.9 no.2
• /
• pp.35-41
• /
• 2021

Purpose : It is necessary to demonstrate the effect of non-invasive and non-pharmacological interventions such as manual therapy and therapeutic exercise for the management of nonspecific neck pain. In the present study, we aimed to investigate the efficacy of myofascial release therapy plus therapeutic exercise for disability owing to neck pain and quality of life in individuals with nonspecific neck pain. Methods : Eighteen participants with nonspecific neck pain were randomly allocated to intervention (n=9) and control groups (n=9). The intervention group received a myofascial release therapy for 20 min and performed neck stabilization exercises for 30 min twice a week for 4 weeks. The control group performed neck stabilization exercises for 30 min twice a week for 4 weeks at the same time points as the intervention group. Disability owing to neck pain and quality of life were quantified using the neck disability index (NDI) and the Korean version of the World Health Organization Quality of Life Brief Version (WHOQOL-BREF), respectively. NDI and WHOQOL-BREF were assessed before and after intervention. Results : The disability owing to neck pain significantly changed between the groups over time (total score of NDI, p=.049). There were significant time and group interactions in pain (pain intensity of NDI, p=.035) and concentration (concentration of NDI, p=.049). Personal care, lifting, reading, headaches, work, driving, sleeping, and recreation did not show significant improvement between the groups over time. Total score, overall quality of life and general health, physical health domain, psychological domain, social relationships domain, and environmental domain quantified by WHOQOL-BREF did not show significant improvements between the groups over time. Conclusion : These results suggest the clinical use of myofascial release therapy in addition to therapeutic exercise for the management of nonspecific neck pain. Further studies are needed to generalize the findings of this study.

• Journal of Trauma and Injury
• /
• v.35 no.1
• /
• pp.3-11
• /
• 2022

Purpose: The increasing use of electric personal mobility devices (ePMDs) has been accompanied by an increasing incidence of associated accidents. This study aimed to investigate the characteristics of ePMD-related injuries and their associated factors. Methods: This cross-sectional study was conducted using data from the Emergency Department-based Injury In-depth Surveillance database from 2014 to 2018. All patients who were injured while operating an ePMD were eligible. The primary outcome was the rate of severe injury, defined as an excess mortality ratio-adjusted Injury Severity Score of ≥25. We calculated the adjusted odds ratios (AORs) of outcomes associated with ePMD-related injuries. Results: Of 1,391,980 injured patients, 684 (0.05%) were eligible for inclusion in this study. Their median age was 28 years old, and most injuries were sustained by men (68.0%). The rate of ePMD-related injuries increased from 3.1 injuries per 100,000 population in 2014 to 100.3 per 100,000 population in 2018. A majority of the injuries occurred on the street (32.7%). The most commonly injured area was the head and face (49.6%), and the most common diagnosis was superficial injuries or contusions (32.9%). Being aged 55 years or older (AOR, 3.88; 95% confidence interval, 1.33-11.36) and operating an ePMD while intoxicated (AOR, 2.78; 95% confidence interval, 1.52-5.08) were associated with severe injuries. Conclusions: The number of emergency room visits due to ePMD-related injuries is increasing. Old age and drunk driving are both associated with serious injuries. Active traffic enforcement and safety regulations regarding ePMDs should be implemented to prevent severe injuries caused by ePMD-related accidents.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.1099-1110
• /
• 2023

The recognition system of autonomous driving and robot navigation performs vision work such as object recognition, tracking, and lane detection after multi-sensor fusion to improve performance. Currently, research on a deep learning model based on the fusion of a camera and a lidar sensor is being actively conducted. However, deep learning models are vulnerable to adversarial attacks through modulation of input data. Attacks on the existing multi-sensor-based autonomous driving recognition system are focused on inducing obstacle detection by lowering the confidence score of the object recognition model.However, there is a limitation that an attack is possible only in the target model. In the case of attacks on the sensor fusion stage, errors in vision work after fusion can be cascaded, and this risk needs to be considered. In addition, an attack on LIDAR's point cloud data, which is difficult to judge visually, makes it difficult to determine whether it is an attack. In this study, image scaling-based camera-lidar We propose an attack method that reduces the accuracy of LCCNet, a fusion model (camera-LiDAR calibration model). The proposed method is to perform a scaling attack on the point of the input lidar. As a result of conducting an attack performance experiment by size with a scaling algorithm, an average of more than 77% of fusion errors were caused.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.1119-1127
• /
• 2009

In PGA golf, total prize money and average score are good response variable related to golf skills such as driving distance, green in regulation and putts per green in regulation. But it's not easy to predict the winner of coming tournament. Thus I applied Neural Networks which has pretty good advantages for non-linear complex modeling to binary data. In neural network architectures, I applied NRBF and MLP architecture model for binary data which represent who had a win or not.

• Journal of Distribution Science
• /
• v.15 no.11
• /
• pp.49-57
• /
• 2017

Purpose - In Korea, general super markets have a great impact on the market performance of traditional markets. We propose a modified two stage DEA model for evaluating the performance of traditional markets in Incheon, Korea by identifying the influence of external environmental factors including the presence of general super markets as non-discretionary variables in DEA. Research design, data, and methodology - After obtaining bias-corrected estimates of original DEA efficiency scores using the input and output data of 49 traditional markets, we regress them on several external environmental factors by bootstrap-truncated regression. Results - We obtain bias-corrected efficiency scores from the original DEA efficiency scores by bootstrap and among the five environmental factors, the residential population and the presence of general super markets or SSMs can be considered as the driving forces influencing bias-corrected efficiency scores, positively and negatively, respectively. Conclusions - When DEA efficiency scores tend to be overestimated, we need to use a biased-corrected efficiency score by bootstrap. It is important to note that the efficiency of traditional markets can be largely influenced by external environmental factors such as the presence of general super markets or SSMs that traditional markets can not control. Therefore, it is desirable to consider such environmental factors appropriately for a reasonable performance evaluation.

• Physical Therapy Korea
• /
• v.17 no.3
• /
• pp.68-76
• /
• 2010

The purpose of this study was to establish reliability and validity of the Northwick Park Neck Pain Questionnaire (NPQ) translated into Korean for neck pain patients. Sixty-two subjects (35 males, 27 females) with neck pain enrolled in the study. They completed a standardized self-administered questionnaire that included pain intensity, sleeping, sensory at night, duration of symptoms, carrying, reading and watching television, working, social activities, and driving. Reliability was determined by intraclass correlation coefficient (ICC) and Cronbach's alpha by internal consistency. Validity was examined by correlating the NPQ scores to the Visual Analog Scale (VAS) score. Test-retest reliability of the translated versions of the NPQ was good ICC(2,1)=.83 (95%CI=.85~.95). Cronbach's alpha value for NPQ was found to be .87 and this was statistically significant (p<.05). The criterion-related validity coefficients was .75 (p<.01). We concluded that the Korean version NPQ was shown to be a reliable and valid instrument for the assessment of neck pain.