• Publications of The Korean Astronomical Society
• /
• v.18 no.1
• /
• pp.37-41
• /
• 2003

We have developed a seeing monitoring system and measured seeing variation of the Bohyunsan Optical Astronomy Observatory (BOAO) and the Sobaeksan Optical Astronomy Observatory (SOAO) using a small telescope system. Our seeing monitoring system is similar to the differential image motion monitor (DIMM) installed at the ESO. The ooly difference between the BOAO and the SOAO seeing monitoring system is a detector system, a video camera at the BOAO and ST-4 camera at the SOAO. We confirmed that the seeing monitoring system at the SOAO can measure average seeing size inspite of its simple detector system. From the BOAO seeing measurement, we found that the seeing size changes fast. We expect that our seeing monitoring system could be used for real time seeing monitoring after some improvement, and the data to be obtained would be very useful when we build adaptive optic system in the future.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.3
• /
• pp.239-248
• /
• 2004

Optical coherence tomography(OCT) is high resolution imaging system which can see the cross section of microscopic organs in the living tissue. In this paper, we analyze the relation between the light source and the resolution of modulated signal in Michelson interferometer. We construct 1-D OCT signal processing hardware such as amplifiers, filters, and demodulate electronic signals from the photo detector. In order to get 2-D OCT image, the synchronization among optical delay line, sample stage and A/D converter is dealt with. In experiments, we verify analog and digital signal processing blocks which apply to the stacks of glasses. Finally we aquire high resolution 2-D OCT image with respect to the onion tissue. We expect that this result can be applied to the medical instrument through performance improvement.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.3
• /
• pp.369-376
• /
• 2014

Recent climate change and ensure each country's marine fisheries resources due to the sharp decline to address the eco-friendly farming has emerged. Marine aquaculture operations to provide ease of use for the fishing ship cranes. In this paper, we use the default handler for the shipping cranes and improve the working environment of the received video equipment in the work area through the monitoring and analysis of visual information to optimize the image, convenient and high visibility for workers to have real-time video guide system is proposed.

• Journal of Korean Society of Transportation
• /
• v.29 no.2
• /
• pp.111-121
• /
• 2011

This paper discusses a method for optimizing the semi-actuated traffic signal control system by adjusting the initial interval according to the number of vehicles waiting for the green light in the actuated phase. We also present a Left-Turn actuated traffic signal control strategy that examines the vehicular noise in the detection area and determines the phase extension and the gap-out. In order to detect the vehicles in real-time, an image detector's Video Image Tracking technology was adopted. A 'Zone in Zone'method was implemented, and the image detection area is segmented into three zones: 1) Zone1 for verifying a vehicles obligatory presence, 2) Zone2 for counting the standby vehicles, and 3) Zone3 for examining the number of vehicles that have passed. The on-site assessment of the Left Turn Actuated Control is carried out using CORSIM, and the results show that the Control Delay decreased by 23.10%, 15.06%, and 4.34% compared to the delays resulted from pre-timed control, semi-actuated control-1 and semi-actuated control-2 traffic signal control systems respectively. The Queue Time also decreased by 36.24%, 20.10% and the Total Time by 14.36%, 7.02% for the same scenario. Which clearly demonstrates the operational efficiency. A sensitivity analysis reveals that the improvement from the propose traffic control strategy tends to increase as the through traffic volume reaches a saturated condition and the left-turn traffic volume decreases.

• Journal of KIISE
• /
• v.43 no.11
• /
• pp.1281-1297
• /
• 2016

In recent years, personal videos have seen a tremendous growth due to the substantial increase in the use of smart devices and networking services in which users create and share video content easily without many restrictions. However, taking both into account would significantly improve event detection performance because videos generally have multiple modalities and the frame data in video varies at different time points. This paper proposes an event detection method. In this method, high-level features are first extracted from multiple modalities in the videos, and the features are rearranged according to time sequence. Then the association of the modalities is learned by means of DNN to produce a personal video event detector. In our proposed method, audio and image data are first synchronized and then extracted. Then, the result is input into GoogLeNet as well as Multi-Layer Perceptron (MLP) to extract high-level features. The results are then re-arranged in time sequence, and every video is processed to extract one feature each for training by means of DNN.

• Journal of Korean Society of Transportation
• /
• v.14 no.2
• /
• pp.89-118
• /
• 1996

The development and implementation of a real-time, traffic adaptive control scheme based on fuzzy logic through Video Image Detector systems (VIDs) is presented. Through VIDs based image processing, fuzzy logic can be used for a real-time traffic adaptive signal control scheme. Fuzzy control logic allows linguistic and inexact traffic data to be manipulated as a useful tool in designing signal timing plans. The fuzzy logic has the ability to comprehend linguistic instructions and to generate control strategy based on a priori verbal communication. The implementation of fuzzy logic controller for a traffic network is introduced. Comparisons are made between implementations of the fuzzy logic controller and the actuated controller in an isolated intersection. The results obtained from the application of the fuzzy logic controller are also compared with those corresponding to a pretimed controller for the coordinated intersections. Simulation results from the comparisons indicate the performance of the system is between under the fuzzy logic controller. Integration of the aforementioned schemes into and ATMS framework will lead to real-time adjustment of the traffic control signals, resulting in significant reduction in traffic congestion.

• Journal of the Korea Convergence Society
• /
• v.10 no.10
• /
• pp.15-21
• /
• 2019

X-rays with shorter wavelengths than ultraviolet light have very good penetration power. It is convergence in industrial and medical fields has been used a lot. n particular, in the industrial field, various researches have been conducted on the detection of foregin body inside metal that can occur in the production process of products such as metal using x-ray, a non-destructive inspection device. Detectors are becoming increasingly popular for the popularization of DR (Digital Radiography) photography methods that digitally acquire X-ray video images. However, there are cases where foreign body detection is impossible depending on the sensor noise and sensitivity inside the detector. When producing a metal product, since the defective rate of the produced product may increase due to contamination of the foreign body, accurate detection is necessary. In this paper, we provide a correction model for X-ray images acquired in order to improve the efficiency of defect detection such as foreign body inside metal. When applied to defect detection in the production process of metal products through the proposed model, it is expected that the detection of product defects can be processed accurately and quickly.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Smart Structures and Systems
• /
• v.23 no.4
• /
• pp.359-371
• /
• 2019

As part of a structural health monitoring system, the relative geometric relationship between a ship and bridge has been recognized as important for bridge authorities and ship owners to avoid ship-bridge collision. This study proposes a novel computer vision method for the real-time geometric parameter identification of moving ships based on a single shot multibox detector (SSD) by using transfer learning techniques and monocular vision. The identification framework consists of ship detection (coarse scale) and geometric parameter calculation (fine scale) modules. For the ship detection, the SSD, which is a deep learning algorithm, was employed and fine-tuned by ship image samples downloaded from the Internet to obtain the rectangle regions of interest in the coarse scale. Subsequently, for the geometric parameter calculation, an accurate ship contour is created using morphological operations within the saturation channel in hue, saturation, and value color space. Furthermore, a local coordinate system was constructed using projective geometry transformation to calculate the geometric parameters of ships, such as width, length, height, localization, and velocity. The application of the proposed method to in situ video images, obtained from cameras set on the girder of the Wuhan Yangtze River Bridge above the shipping channel, confirmed the efficiency, accuracy, and effectiveness of the proposed method.

• International Journal of Computer Science & Network Security
• /
• v.22 no.11
• /
• pp.121-126
• /
• 2022

The number of vehicles has increased exponentially over the past 20 years due to technological advancements. It is becoming almost impossible to manually control and manage the traffic in a city like Karachi. Without license plate recognition, traffic management is impossible. The Framework for License Plate Detection & Recognition to overcome these issues is proposed. License Plate Detection & Recognition is primarily performed in two steps. The first step is to accurately detect the license plate in the given image, and the second step is to successfully read and recognize each character of that license plate. Some of the most common algorithms used in the past are based on colour, texture, edge-detection and template matching. Nowadays, many researchers are proposing methods based on deep learning. This research proposes a framework for License Plate Detection & Recognition using a custom YOLOv5 Object Detector, image segmentation techniques, and Tesseract's optical character recognition OCR. The accuracy of this framework is 0.89.