• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.227-241
• /
• 2021

Object detection techniques based on deep learning such as YOLO have high detection performance and precision in a single channel video stream. In order to expand to multiple channel object detection in real-time, however, high-performance hardware is required. In this paper, we propose a novel back-end server framework, a real-time AI vision platform (RAVIP), which can extend the object detection function from single channel to simultaneous multi-channels, which can work well even in low-end server hardware. RAVIP assembles appropriate component modules from the RODEM (real-time object detection module) Base to create per-channel instances for each channel, enabling efficient parallelization of object detection instances on limited hardware resources through continuous monitoring with respect to resource utilization. Through practical experiments, RAVIP shows that it is possible to optimize CPU, GPU, and memory utilization while performing object detection service in a multi-channel situation. In addition, it has been proven that RAVIP can provide object detection services with 25 FPS for all 16 channels at the same time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2148-2161
• /
• 2019

In this paper, we present a real-time cattle action recognition algorithm to detect the estrus phase of cattle from a live video stream. In order to classify cattle movement, specifically, to detect the mounting action, the most observable sign of the estrus phase, a simple yet effective feature description exploiting motion history images (MHI) is designed. By learning the proposed features using the support vector machine framework, various representative cattle actions, such as mounting, walking, tail wagging, and foot stamping, can be recognized robustly in complex scenes. Thanks to low complexity of the proposed action recognition algorithm, multiple cattle in three enclosures can be monitored simultaneously using a single fisheye camera. Through extensive experiments with real video streams, we confirmed that the proposed algorithm outperforms a conventional human action recognition algorithm by 18% in terms of recognition accuracy even with much smaller dimensional feature description.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.45 no.3
• /
• pp.177-187
• /
• 2009

This paper describes a automatic video surveillance system(AVSS) with long range and 360$^{\circ}$ coverage that is automatically rotated in an elevation over azimuth mode in response to the TTM(tracked target message) signal of vessels tracked by ARPA(automatic radar plotting aids) radar. This AVSS that is a video security and tracking system supported by ARPA radar, CCTV(closed-circuit television) camera system and other sensors to automatically identify and track, detect the potential dangerous situations such as collision accidents at sea and berthing/deberthing accidents in harbor, can be used in monitoring the illegal fishing vessels in inshore and offshore fishing ground, and in more improving the security and safety of domestic fishing vessels in EEZ(exclusive economic zone) area. The movement of the target vessel chosen by the ARPA radar operator in the AVSS can be automatically tracked by a CCTV camera system interfaced to the ECDIS(electronic chart display and information system) with the special functions such as graphic presentation of CCTV image, camera position, camera azimuth and angle of view on the ENC, automatic and manual controls of pan and tilt angles for CCTV system, and the capability that can replay and record continuously all information of a selected target. The test results showed that the AVSS developed experimentally in this study can be used as an extra navigation aid for the operator on the bridge under the confusing traffic situations, to improve the detection efficiency of small targets in sea clutter, to enhance greatly an operator s ability to identify visually vessels tracked by ARPA radar and to provide a recorded history for reference or evidentiary purposes in EEZ area.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.1
• /
• pp.1-6
• /
• 2021

Recently, unmanned facilities have been introduced and operated in a way that reduces the cost and development of IT technology. Although unmanned facilities have advantages in terms of efficiency and economy, they have disadvantages such as failure of unmanned facilities and malfunctions, causing damage to facilities caused by intruders, and information leakage. In addition, it is necessary to visit the person in charge at all times to inspect the unmanned facilities, resulting in management costs. In this paper, we designed a system that checks the status of unmanned surveillance facilities in real time to check and automatically recover problems such as malfunctions, and to notify managers of situations by text messages in real time. The system to be designed consists of an integrated network video server (NVR) that receives and determines information on the operation status of the main equipment such as video, sound, and lighting, and a real-time text message using an SMS server.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.1
• /
• pp.175-186
• /
• 2018

Due to global warming, various natural disasters such as floods and localized heavy rains are increasing. If a natural disaster can be detected and analyzed in advance and effectively, it can prevent enormous damage due to natural disasters. Recent development in visual sensor technologies has encouraged various studies on monitoring environments including rivers. In this paper, we propose a method to detect river regions from river images which can be exploited for river surveillance systems using video sensor networks. In the proposed method, we first analyze the color properties of the river region and the background region of a image and then propose a way to select the proper color channel and binarize the image to detect the river region. It is shown by experimental results that the proposed method is simple but detects river regions accurately.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.4
• /
• pp.95-103
• /
• 2009

Global interests for construction of 3 dimensional spatial information has risen due to development of measurement sensors and data processing technologies. In spite of criticism for the violation of personal privacy, CCTV cameras equipped in outdoor public space of urban area are used as a fundamental sensor for traffic management, crime prevention or hazard monitoring. For safety guarantee in urban environment and disaster prevention, a surveillance system integrating pre-constructed 3 dimensional spatial information with CCTV data or video sequence is needed for monitoring and observing emergent situation interactively in real time. In this study, we proposed applicability of the prototype system for web visualization based on X3D, an international standard of real time web visualization, by integrating 3 dimensional spatial information with video sequence.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.2
• /
• pp.209-214
• /
• 2018

The subway, which is an urban railway, is the core of public transportation. Urban railways are always exposed to serious problems such as theft, crime and terrorism, as many passengers use them. Especially, due to the nature of urban railway environment, the scope of surveillance is widely dispersed and the range of surveillance target is rapidly increasing. Therefore, it is difficult to perform comprehensive management by passive surveillance like existing CCTV. In this paper, we propose the implementation, design method and object recognition algorithm for intelligent object recognition system in urban railway. The object recognition system that we propose is to analyze the camera images in the history and to recognize the situations where there are objects in the landing area and the waiting area that are not moving for more than a certain time. The proposed algorithm proved its effectiveness by showing detection rate of 100% for Selected area detection, 82% for detection in neglected object, and 94% for motionless object detection, compared with 84.62% object recognition rate using existing Kalman filter.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.4
• /
• pp.109-117
• /
• 2019

In modern society, due to various accidents and crime threats committed to an unspecified number of people, individual security awareness is increasing throughout society and various surveillance techniques are being actively studied. Still, there is a decline in robustness due to many problems, requiring higher reliability monitoring techniques. Thus, this paper suggests a real-time change detection technique to complement the low robustness problem in various environments and dynamic/static change detection and to solve the cost efficiency problem. We used the Self-Organizing Map (SOM) applied as a data clustering technique to implement change detection, and we were able to confirm the superiority of noise robustness and abnormal detection judgment compared to the detection technique applied to the existing image surveillance system through simulation in the indoor office environment.

• Journal of information and communication convergence engineering
• /
• v.12 no.3
• /
• pp.168-172
• /
• 2014

In this paper, we propose an algorithm that can efficiently detect and monitor multiple ships in real-time. The proposed algorithm uses morphological operations and edge information for detecting and tracking ships. We used smoothing filter with a $3{\times}3$ Gaussian window and luminance component instead of RGB components in the captured image. Additionally, we applied Sobel operator for edge detection and a threshold for binary images. Finally, object labeling with connectivity and morphological operation with open and erosion were used for ship detection. Compared with conventional methods, the proposed method is meant to be used mainly in coastal surveillance systems and monitoring systems of harbors. A system based on this method was tested for both stationary and non-stationary backgrounds, and the results of the detection and tracking rates were more than 97% on average. Thousands of image frames and 20 different video sequences in both online and offline modes were tested, and an overall detection rate of 97.6% was achieved.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.9
• /
• pp.120-125
• /
• 2016

For the purpose of protecting personal property and lives from incidents, accidents, and threats such as terrorism, video surveillance equipment has been installed and operates in many places. Video surveillance technology has gradually developed into high-quality, high-definition equipment, and a lot of products have been launched. However, closed circuit television (CCTV) equipment for security purposes can invade a person's privacy. In this paper, we propose a way to protect personal video images using meta-data in an intelligent Internet protocol (IP) camera. We designed the system to mask personal video information from meta-data, define the method of image-information access according to user privileges, and show how to utilize the meta-data during storage and recorded data searches. The suggested system complies with guidelines for CCTV installation and operation from Korea's Ministry of the Interior. Installed on only a single server so far, due to the limitations and technical difficulties of hardware performance, it has been difficult to find a method that can be applied to personal image information using real-time protection techniques. Applying the method proposed in this paper can satisfy the guidelines, reduce server costs, and reduce system complexity.