• International journal of advanced smart convergence
• /
• v.10 no.4
• /
• pp.110-116
• /
• 2021

This paper proposed a real-time earlobe detection system using deep learning on the web. Existing deep learning-based detection methods often find independent objects such as cars, mugs, cats, and people. We proposed a way to receive an image through the camera of the user device in a web environment and detect the earlobe on the server. First, we took a picture of the user's face with the user's device camera on the web so that the user's ears were visible. After that, we sent the photographed user's face to the server to find the earlobe. Based on the detected results, we printed an earring model on the user's earlobe on the web. We trained an existing YOLO v5 model using a dataset of about 200 that created a bounding box on the earlobe. We estimated the position of the earlobe through a trained deep learning model. Through this process, we proposed a real-time earlobe detection system on the web. The proposed method showed the performance of detecting earlobes in real-time and loading 3D models from the web in real-time.

• Journal of the Semiconductor & Display Technology
• /
• v.9 no.2
• /
• pp.79-85
• /
• 2010

In this paper, we propose a web inspection system (WIS) for real-time detection of paper defects which can cause critical fractures during papermaking process. Our system incorporates high speed line-scan camera, lighting system, and detection algorithm to provide robust and precise detection of paper defects in real-time. Since edge defects are very crucial to the paper fractures, our system focuses on the edge region of the paper instead of inspecting the whole paper area. In our algorithm, image projection and sub-pixel operation are utilized to detect the edge defects precisely and connected component labeling and shape analysis techniques are adopted to extract various kinds of the region defects. Experimental results revealed that our web inspection system is very efficient for detecting paper defects during papermaking processes.

• Journal of Microbiology
• /
• v.41 no.4
• /
• pp.320-326
• /
• 2003

In order to identify Vibrio vulnificus in the Yellow Sea near Gunsan, Korea during the early and late summers, the efficiency of the real-time quantitative TaqMan PCR was compared to the efficiency of the conventional PCR and Biolog identification system^TM. Primers and a probe were designed from the hemolysin/cytolysin gene sequence of V. vulnificus strains. The number of positive detections by real-time quantitative TaqMan PCR, conventional PCR, and the Biolog identification system from seawater were 53 (36.8%), 36 (25%), and 10 strains (6.9%), respectively, among 144 samples collected from Yellow Sea near Gunsan, Korea. Thus, the detection method of the real-time quantitative TaqMan PCR assay was more effective in terms of accuracy than that of the conventional PCR and Biolog system. Therefore, our results showed that the real-time TaqMan probe and the primer set developed in this study can be applied successfully as a rapid screening tool for the detection of V. vulnificus.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.1
• /
• pp.1-8
• /
• 2022

In this paper, we propose a fire detection system based on CCTV images using an object tracking technology with YOLOv4 model capable of real-time object detection and a DeepSORT algorithm. The fire detection model was learned from 10800 pieces of learning data and verified through 1,000 separate test sets. Subsequently, the fire detection rate in a single image and fire detection maintenance performance in the image were increased by tracking the detected fire area through the DeepSORT algorithm. It is verified that a fire detection rate for one frame in video data or single image could be detected in real time within 0.1 second. In this paper, our AI fire detection system is more stable and faster than the existing fire accident detection system.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.10
• /
• pp.125-133
• /
• 2016

In this paper, we propose a real-time fault monitoring and dual system design of the countdown time-generating system, which is the main component of the mission control system. The countdown time-generating system produces a countdown signal that is distributed to mission control system devices. The stability of the countdown signal is essential for the main launch-related devices because they perform reserved functions based on the countdown time information received from the countdown time-generating system. Therefore, a reliable and fault-tolerant design is required for the countdown time-generating system. To ensure system reliability, component devices should be redundant and faults should be monitored in real time to manage the device changeover from Active mode to Standby mode upon fault detection. In addition, designing different methods for mode changeover based on fault classification is necessary for appropriate changeover. This study presents a real-time fault monitoring and changeover system, which is based on the dual system design of countdown time-generating devices, as well as experiment on real-time fault monitoring and changeover based on fault inputs.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.428-442
• /
• 2022

A reliable automatic passenger counting (APC) system is a key point in transportation related to the efficient scheduling and management of transport routes. In this study, we introduce a lightweight head detection network using deep learning applicable to an embedded system. Currently, object detection algorithms using deep learning have been found to be successful. However, these algorithms essentially need a graphics processing unit (GPU) to make them performable in real-time. So, we modify a Tiny-YOLOv3 network using certain techniques to speed up the proposed network and to make it more accurate in a non-GPU environment. Finally, we introduce an APC system, which is performable in real-time on embedded systems, using the proposed head detection algorithm. We implement and test the proposed APC system on a Samsung ARTIK 710 board. The experimental results on three public head datasets reflect the detection accuracy and efficiency of the proposed head detection network against Tiny-YOLOv3. Moreover, to test the proposed APC system, we measured the accuracy and recognition speed by repeating 50 instances of entering and 50 instances of exiting. These experimental results showed 99% accuracy and a 0.041-second recognition speed despite the fact that only the CPU was used.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.118.3-118
• /
• 2002

$\textbullet$ Introduction $\textbullet$ The Structure and Operation of an Electronic Jacquard $\textbullet$ Design of Real-Time Error Detection System $\textbullet$ The System Realization and Verification $\textbullet$ Conclusion

• Journal of Food Hygiene and Safety
• /
• v.34 no.3
• /
• pp.290-295
• /
• 2019

Salmonella spp. are frequently associated with food and are among the most important foodborne pathogens. The recent Salmonella out breaks in Korea was associated with chocolate mousse cakes served with school meals during September 2018. The objective of this research was to compare the 3M Molecular Detection Assay 2 - Salmonella and the Korean Standard Method of Salmonella in artificially inoculated mousse (chocolate and cheese) and tiramisu cakes. Mousse (chocolate and cheese) and tiramisu cakes were artificially inoculated with S. Typhimurium. Twenty five gram of sample was enriched with 225 mL buffered peptone water for incubation at $37^{\circ}C$ for 24 h. After enrichment, the cultures were analyzed by using the 3M Molecular Detection Assay 2 - Salmonella and the Korean Standard Method. Most of the inoculated samples showed similar results except the chocolate mousse cakes, in which real-time PCR was unable to detect S. Typhimurium even after $10^4CFU/25g$ of inoculation. However, S. Typhimurium inoculated at a concentration of $10^0CFU/25g$ was detected by using 3M Molecular Detection Assay 2 - Salmonella. In chocolate mousse, detection of S. Typhimurium using real-time PCR was partially successful when dark chocolate was added at less than 15%. Negative results in real-time PCR and 3M Molecular Detection Assay 2 - Salmonella were confirmed by gel electrophoresis. The data indicated that dark chocolate could inhibit amplification of the target gene in the PCR reactions. In conclusion, the 3M Molecular Detection Assay 2 - Salmonella was better than the Korean Standard Method (real-time PCR) for the detection of S. Typhimurium in chocolate mousse cakes and chocolate mousse.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.24 no.2
• /
• pp.127-136
• /
• 2016

The camera based object detection systems should satisfy the recognition performance as well as real-time constraints. Particularly, in safety-critical systems such as Autonomous Emergency Braking (AEB), the real-time constraints significantly affects the system performance. Recently, multi-core processors and system-on-chip technologies are widely used to accelerate the object detection algorithm by distributing computational loads. However, due to the advanced hardware, the complexity of system architecture is increased even though additional hardwares improve the real-time performance. The increased complexity also cause difficulty in migration of existing algorithms and development of new algorithms. In this paper, to improve real-time performance and design complexity, a task scheduling strategy is proposed for visual object tracking systems. The real-time performance of the vision algorithm is increased by applying pipelining to task scheduling in a multi-core processor. Finally, the proposed task scheduling algorithm is applied to crosswalk detection and tracking system to prove the effectiveness of the proposed strategy.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1997.11a
• /
• pp.417-433
• /
• 1997

A real-time operating system has focused primary on techniques to minimize processing time, with a secondary emphasis on system reliability issues. Conversely, fault-tolerant system has concentrated on using recourse and information redundancy to maximize the availability and reliability of the system, with a lesser emphasis on performance. We have developed a fault-tolerant and real-time operations system which support a powerful concurrent runtime environment under the above requirements. In this paper, we present an overview of real-time systems, design and implementation of a duplex architecture using advanced concepts and technologies such as fast " fault detection", "fault isolation" and "fault recovery" Because the duplex architecture has two dentical hardware elements and has several recovery steps and hierarchy to facilitate a fast recovery which must be proceeded by a prompt fault detection and isolation. Thus it makes possible to minimize the overhead of the systems including hardware and software and guarantee the service continuity of he systems.