• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.546-554
• /
• 2019

Typically, closed-circuit television (CCTV) monitoring is mainly used for post-processes (i.e. to provide evidence after an incident has occurred), but by using a streaming video feed, machine-based learning, and advanced image recognition techniques, current technology can be extended to respond to crimes or reports of missing persons in real time. The multi-CCTV cooperation technique developed in this study is a program model that delivers similarity information about a suspect (or moving object) extracted via CCTV at one location and sent to a monitoring agent to track the selected suspect or object when he, she, or it moves out of range to another CCTV camera. To improve the operating efficiency of local government CCTV control centers, we describe here the partial automation of a CCTV control system that currently relies upon monitoring by human agents. We envisage an integrated crime prevention service, which incorporates the cooperative CCTV network suggested in this study and that can easily be experienced by citizens in ways such as determining a precise individual location in real time and providing a crime prevention service linked to smartphones and/or crime prevention/safety information.

• The Journal of the Korea Contents Association
• /
• v.6 no.12
• /
• pp.1-11
• /
• 2006

Many research demonstrations and commercial applications have been tried to develop face detection and recognition systems. Human face detection plays an important role in applications such as access control and video surveillance, human computer interface, identity authentication, etc. There are some special problems such as a face connected with background, faces connected via the skin color, and a face divided into several small parts after skin region segmentation in generally. It can be allowed many face detection techniques to solve the first and second problems. However, it is not easy to detect a face divided into several parts of regions for reason of different illumination conditions in the third problem. Therefore, we propose an efficient modified skin segmentation algorithm to solve this problem because the typical region segmentation algorithm can not be used to. Our algorithm detects skin regions over the entire image, and then generates face candidate regions using our skin segmentation algorithm For each face candidate, we implement the procedure of region merging for divided regions in order to make a region using adjacency between homogeneous regions. We utilize various different searching window sizes to detect different size faces and a face detection classifier based on a back-propagation algorithm in order to verify whether the searching window contains a face or not.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.3
• /
• pp.258-268
• /
• 2001

The compression method is necessarily used to send the high quality moving picture that contains a number of data in image processing. In the field of moving picture compression method, the motion estimation algorithm is used to reduce the temporal redundancy. Block matching algorithm to be usually used is distinguished partial search algorithm with full search algorithm. Full search algorithm be used in this paper is the method to compare the reference block with entire block in the search window. It is very efficient and has simple data flow and control circuit. But the bigger the search window, the larger hardware size, because large computational operation is needed. In this paper, we design the full search block matching motion estimator. Using the DCT DC values, we decide luminance. And we apply 3 bit compare-selector using bit plane to I(Intra coded) picture, not using 8 bit luminance signals. Also it is suggested that use the same selective bit for the P(Predicted coded) and B(Bidirectional coded) picture. We compare based full search method with PSNR(Peak Signal to Noise Ratio) for C language modeling. Its condition is the reference block 8$\times$8, the search window 24$\times$24 and 352$\times$288 gray scale standard video images. The result has small difference that we cannot see. And we design the suggested motion estimator that hardware size is proved to reduce 38.3％ for structure I and 30.7％ for structure II. The memory is proved to reduce 31.3％ for structure I and II.

• Journal of the Korean Society of Floral Art and Design
• /
• no.41
• /
• pp.93-114
• /
• 2019

The purpose of this study is to analyze the characteristics and problems of the content of flower design education videos on YouTube and to search for improvement direction. The subjects of analysis were 129 pieces of videos uploaded in the last one year including 'craftman floral design' as a search term. The result shows that contents covered were practical lectures, theory lectures, test related tips, job and character introduction, test work, educational guidance and publicity. The production format could be divided into studio lecture, classroom lecture, video feature, interview, Vlog, and television program. The hub-type programming strategy that periodically uploads the videos satisfying the target audiences' interests is mostly applied. The type of lecture covered 'practical skill test' got a good response from the users. Overall, content diversity, interaction between creators and users, and harmonious programming strategies are lacking. In order to improve this, it is necessary for emotional and expressive creators to pioneer differentiated fields and practice based on actual field. The introduction of interactive elements such as games and quizzes and the application of new media technologies such as VR and AR are worth trying. Three strategic types of 'hero', 'hub', and 'how to' should be applied complementary. As the demand for education content related to flower design is expected to expand in the future, it is required to develop content that can be used in various platforms, foster professional creators, and develop associated business models.

• Journal of Industrial Convergence
• /
• v.22 no.10
• /
• pp.21-28
• /
• 2024

Personal Mobility devices pose a significant safety risk to users depending on road conditions. With the recent surge in usage, the number of accidents has also increased, highlighting the need for a preventative system. This study aims to design and develop a warning system to enhance the safety of drivers by detecting road obstacles. To achieve this, a dataset is constructed using images of road obstacles, and the YOLOv5 model is trained. The system is based on the Raspberry Pi 4B, which processes video frames captured by a camera in real-time and triggers an LED warning when an obstacle is detected. The Flask framework is used to monitor the obstacle detection status in real time. Additionally, a GPS sensor is utilized to collect the user's location and speed data, and an auditory warning is triggered via a buzzer if the set speed is exceeded. In the future, this system could be expanded to transmit detected road obstacles and GPS information to a server, providing users with real-time road safety information. The results of this study can serve as essential technology for developing such a system.

• Journal of Broadcast Engineering
• /
• v.9 no.1
• /
• pp.83-90
• /
• 2004

DVB-MHP recommends that NPT(normal play time) be used as the times of stream events. NPT is the local time within an event(TV program). But we found that commercial transport stream (TS) generators and middlewares for DVB-MHP settop boxes are not ready to support the use of NPT by applications. In particular, TS generators do not create NPT reference descriptors needed to reconstruct NPT at the TV receiver. This situation is undesirable because program providers cannot experiment with the idea of synchronized applications. So we have implemented a TS generator that inserts NPT reference descriptors to TS and MyGetNPT API to approximately reconstruct NPT. STC (system time clock) is needed to reconstruct NPT, but Xlets are not allowed to read it. So, we approximate STC by using PCR (program clock reference) and the Java system tune. In this method, the stream generator extrats PCRs from an existing TS and inserts them into null TS packets in the form of MPEG sections, which can be read by Xlets. Because PCRs are displaced into new positions in TS, their values should be adjusted based on the time intervals between the original positions and the new positions. We implemented a synchronized application by using our TS generator and MyGetNPT API, where the task of stream events are to display graphic images. We found that graphic images are displayed where 240 ㎳ from their intended time, where 240ms is a human tolerance for the synchronization skew between graphic image and video.

• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.17 no.1
• /
• pp.28-37
• /
• 2006

Purpose: Evaluation of vocal cord vibration is very important in cases of voice disorders. There are several equipments for examining the vocal fold vibration such as laryngeal stroboscope, ultra high-speed digital imaging system, and videokymograph. Among these, laryngeal stroboscope is the most popular equipment because of easy to examine the laryngeal pathology. However, current laryngo-stroboscopes are too bulky to move and relatively expensive. The purpose of this research is to develope a portable laryngeal stroboscope of equivalent performance with the current equipments. Methods and Materials: Recently developed high luminescent white LEDs(light emitting diodes) are placed at the head of the endoscope as light sources for the CCD image sensor which is also placed at the head with imaging lens. This arrangement eliminates the bulky light source like expensive halogen or xenon lamps as well as the optical light guiding cables. The LEDs are controlled to flash in phase with the voice frequency of the examinee. The CCD captures these strobo images and converts them into video signals for examinations. Results: There was no functional differences between preexisting stroboscope and the newly developed stroboscope of this study. LED light sources and microprocessor based control circuits of the stroboscope enabled the development of flicker-less, hand-held, portable and battery-operating stroboscope. Conclusion: The developed stroboscope is cost-effective, small-sized, easy to use and very easy desirable to bring and to use in any place.

• Journal of Korean Society of Transportation
• /
• v.34 no.3
• /
• pp.207-221
• /
• 2016

In Korea, older pedestrian accounted for 57% of all pedestrian deaths although a ratio of older pedestrian accidents to total pedestrian accidents was only 25.9%. Though ageing population problem becomes more challenging for road safety, little is know about the behaviour of older pedestrian's behaviour. This study aimed to identify road crossing behaviour of older pedestrian at three-lane unsignalized crosswalks using video image analysis and to compare the behaviour of older pedestrian to younger one by indicators including approaching speed, the number of walking steps and other factors. The results showed that there was a difference of approaching time at kerb, waiting time at kerb, the number of glances at kerb, and the number of glances at crossing between two groups under the situation of car approaching to crosswalks. It also showed that older pedestrian usually spent 1.16 times more than younger pedestrian to walk across the crosswalk with only 84.4% of walking speed of younger pedestrian. The number of steps of older pedestrian for road crossing was 1.12 times higher with 90% shorter steps than younger pedestrian. It was concluded that older pedestrian usually decided to walk across in case of 1.67 times longer headway than younger pedestrian's decision. These results could be applied in road and facility design for better safety of older pedestrians.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.7C
• /
• pp.726-732
• /
• 2003

The paper proposes a motion estimation (ME) technique to reduce computational complexity. It is achieved by skipping ME process for macro-blocks decided to be in no need of the operation. Thus, it is called ME skipping technique(MEST). In general, the ME is composed of integer pixel precision ME (IME) followed by half pixel precision ME (HME). The MEST is performed just before an IME process and makes a decision on skipping the IME process according to a criterion based on ME errors of adjacent macro-blocks (MBs) already encoded. When the IME process for a MB is decided to be skipped, which is called ME skip mode, the IME process is skipped and the integer pixel precision motion vector of the MB is just replaced by a predicted vector and used as the input of HME. On the other hands, the IME processes for MBs in ME non-skip mode are not skipped but normally performed. Accordingly, the MEST is very effective to reduce computational complexity when MBs in ME skip mode is abundant. In addition, when the MEST is applied to video encoder, it contributes to more accurate rate control and more robusaess for channel errors. It is experimentally shown that the MEST has the above advantages while maintaining good reconstructed image quality.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.5D
• /
• pp.611-617
• /
• 2009

With advanced technologies applied to transportation, realtime traffic information has been necessary for not only drivers but also agencies. In normal, traffic conditions have been represented to three levels according to congestion: "free", "slow", and "jammed". Those categories and criteria are set up for traffic management even though traffic information is provided for drivers. This study examines how drivers feel current congestion levels and delves into traffic categories and criteria which they recognize. To collect data for drivers' recognition, a survey of freeway travellers is conducted answering the question about traffic flow speed from video image on a freeway section. In the result of the survey, the surveyee preferred a 4-level traffic condition including "delayed" to 3-level traffic condition. As its criteria, 20 km/h, 50 km/h, and 75 km/h were obtained. These results are expected to contribute to building more appropriate traffic information for drivers and providing an operational guideline for Traffic Monitering Centers.