• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1998.06b
• /
• pp.92-97
• /
• 1998

For the efficient searching and browsing of digital video, it is essential to extract the internal structure of the video contents. As an example, a news video consists of several sections such as politics, economics, sports and others, and also each section consists of individual topics. With this information in hand, users can ore easily access the required video frames. This paper addresses the problem of automatic shot boundary detection and selection of representative frames (R-frames), which are the essential step in recognizing the internal structure of video contents. In the shot boundary detection, a new algorithm that have dual detectors which are designed specifically for the abrupt boundaries (cuts) and gradually changing bounaries respectively is proposed. Compared to the existing 미algorithms that mostly have tried to detect both types by a single mechanism, the proposed algorithm is proved to be more robust and accurate. Also in the problem of R-frame selection, simple mechanical approaches such as selecting one frame every other second have been adopted. However this approach often selects too many R-frames in static short, while drops important frames in dynamic shots. To improve the selection mechanism, a new R-frame selection algorithm that uses motion information extracted from pixel difference is proposed.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.5
• /
• pp.13-20
• /
• 2019

Recently, object detection technology based on CNN has been actively studied. Object detection technology is used as an important technology in autonomous vehicles, intelligent image analysis, and so on. In this paper, we propose a rotation change robust object detection system by applying MI-FL (Moment Invariant-Feature Layer) to SSD (Single Shot Multibox Detector) which is one of CNN-based object detectors. First, the features of the input image are extracted based on the VGG network. Then, a total of six feature layers are applied to generate bounding boxes by predicting the location and type of object. We then use the NMS algorithm to get the bounding box that is the most likely object. Once an object bounding box has been determined, the invariant moment feature of the corresponding region is extracted using MI-FL, and stored and learned in advance. In the detection process, it is possible to detect the rotated image more robust than the conventional method by using the previously stored moment invariant feature information. The performance improvement of about 4 ~ 5% was confirmed by comparing SSD with existing SSD and MI-FL.

• International Journal of Internet, Broadcasting and Communication
• /
• v.11 no.4
• /
• pp.76-85
• /
• 2019

We introduce a vision-based object detection method for real-time video surveillance system in low-end edge computing environments. Recently, the accuracy of object detection has been improved due to the performance of approaches based on deep learning algorithm such as Region Convolutional Neural Network(R-CNN) which has two stage for inferencing. On the other hand, one stage detection algorithms such as single-shot detection (SSD) and you only look once (YOLO) have been developed at the expense of some accuracy and can be used for real-time systems. However, high-performance hardware such as General-Purpose computing on Graphics Processing Unit(GPGPU) is required to still achieve excellent object detection performance and speed. To address hardware requirement that is burdensome to low-end edge computing environments, We propose sub-frame analysis method for the object detection. In specific, We divide a whole image frame into smaller ones then inference them on Convolutional Neural Network (CNN) based image detection network, which is much faster than conventional network designed forfull frame image. We reduced its computationalrequirementsignificantly without losing throughput and object detection accuracy with the proposed method.

• Korean Journal of Microbiology
• /
• v.30 no.6
• /
• pp.507-513
• /
• 1992

Serratia marecescens RH1 isolated from soil samples produced large amount of extracellular proteases. One of the genes encoding an extracellular protease form S. marcescens RH1 was cloned in Escherichia coli by shot gun cloning method. The cloned protease, SSP, was stably expressed by its own promoter and excreted into the extracellular medium from E. coli host (ORF) of 3.135 nucleotides corresponding to 1.045 amino acids (112 kDa). The nucleotide and deduced amino acid sequence of SSP showed high overall homology (88%) to one of the S. marcescens protease (27), but low homology to other serine protease families. The optimal pH and temperature of the enzyme were pH 9.0 and 45.deg.C respectively. The activity of protease was inhibited by phenylmethylsulfonyl fluoride (PMSF), which suggests that the enzyme is a serine protease.

• Laser Solutions
• /
• v.17 no.3
• /
• pp.15-18
• /
• 2014

In this paper, we studied a machining method using a diffractive beam splitter (DBS) and multi- pulse repeated radiation for breathable film. We fabricated micro-grooves on polypropylene (PP) films using multi-pulse radiation and one-shot radiation (radiating pulses at once) and a DBS. In the result, width and depth of the PP film using multi-pulse repeated radiation were more precisely controllable. Therefore, this method can be applicable to in manufacturing breathable film precisely at a high speed.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.601-605
• /
• 2009

We have proposed a measurement system for measuring a whole shape of an object easily. The proposed system consists of a camera and a cylinder whose inside is coated by a mirror layer. A target object is placed inside the cylinder and an image is captured by the camera from right above. The captured image includes sets of points that are observed from multiple viewpoints: one is observed directly, and others are observed via the mirror. Therefore, the whole shape of the object can be measured using stereo vision in a single shot. This paper shows that a prototype of the proposed system was implemented and an actual object was measured using the prototype. A method based on a pattern matching which uses a value of SSD (Sum of Squared Difference), and a method based on DP (Dynamic Programming) are employed to identify a set of corresponding points in warped captured images.

• Archives of Reconstructive Microsurgery
• /
• v.7 no.1
• /
• pp.35-40
• /
• 1998

The incidence of brachial plexus injury is increasing because of the development of motor vehicle but the the results of treatment was reported poor due to its complex anatomical structure and changes of function and sensory during the recovery after trauma. But the results of treatment has been improved by the recently introduced high sensitive diagnostic method that can evaluate accurately the site and extent of the injury and treatment method. Restoration of the elbow flexion is the most important goal of treatment after brachial plexus injury and nerve graft, neurotization and muscle transfer were used for methods of treatment. From December 1992 to May 1994, the author performed 6 cases of latissimus dorsi transfer at the same side for the improvement of elbow flexion in the patients of brachial plexus injury. There were 5 cases of male, one case of female and average age was 22 years old. The causes of injury were traffic accident in 3 cases, gun shot injury, falldown and birth injury in each one case and in all cases, the type of injury were upper arm type. The average follow up period were 1 year 5 months ranging from 12 months to 4 years 5 months. In all cases, active elbow flexion was impossible before operation and average muscle power was grade I. We analysed the active range of motion, muscle power and the functional results. At the last follow up, range of active elbow flexion was average $124^{\circ}$ and flexion contracture was average 11 degrees and the average of muscle power was grade IV. In the functional analysis, there were two cases of excellent, three cases of good and 1 case of fair. There was no complications including wound infection, vascular compromise and donor site problem. The results of latissimus dorsi transfer for improvement elbow flexion in the patients of brachial plexus injury is one of the useful mettled for the restoration of elbow flexion.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.1
• /
• pp.96-103
• /
• 2006

This paper proposes a new technique for improving the efficiency of single phase high frequency switch mode boost converter. This converter includes an additional boost converter that follows the main hish frequency switching device. The additional converter, which is controlled at lower frequencies, bypasses almost all the current in the main switch and the high frequency switching loss is greatly reduced. Both switching devices are controlled by a simple method; each controller consists of a one-shot multivibrator, a comparator and an AND gate, and the maximum switching frequency can be limited without any clock generator. The converter works cooperatively in high efficiency and acts as though it were a conventional high frequency switch mode converter with one switching device. This paper describes the proposed converter configuration, design, and discusses the steady state performance concerning the switching loss reduction and efficiency improvement. and the proposed method is verified by computer simulation.

• Convergence Security Journal
• /
• v.22 no.3
• /
• pp.25-32
• /
• 2022

Recently, the image processing industry has been activated as deep learning-based technology is introduced in the image recognition and detection field. With the development of deep learning technology, learning model vulnerabilities for adversarial attacks continue to be reported. However, studies on countermeasures against poisoning attacks that inject malicious data during learning are insufficient. The conventional countermeasure against poisoning attacks has a limitation in that it is necessary to perform a separate detection and removal operation by examining the training data each time. Therefore, in this paper, we propose a technique for reducing the attack success rate by applying modifications to the training data and inference data without a separate detection and removal process for the poison data. The One-shot kill poison attack, a clean label poison attack proposed in previous studies, was used as an attack model. The attack performance was confirmed by dividing it into a general attacker and an intelligent attacker according to the attacker's attack strategy. According to the experimental results, when the proposed defense mechanism is applied, the attack success rate can be reduced by up to 65% compared to the conventional method.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.45 no.1
• /
• pp.63-70
• /
• 2017

In this paper, we design and fabricate a de-orbiter using drag-sail and evaluate deployment characteristics. Without employing an actuator to deploy, the de-orbiter is activated by the SMA wire based the release mechanism and driven by the restoring force of the tape-spring. For efficient storage and deployment of drag-sail, an origami method of original ISO flasher is chosen and low priced mylar film is used as the material of the drag-sail. In addition, through the fault tree analysis method which is one of the one-shot device reliability evaluation methods, we confirm the reliability of the de-orbiter(0.997572) and the Roller failure has the highest criticality. Finally, we find feasibility of the proposed de-orbiter through the deployment demonstration of drag-sail.