• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.2
• /
• pp.40-46
• /
• 2011

This paper proposes a real-time face detection and tracking system that uses neural oscillators which can be applied to access regulation system or control systems of user authentication as well as a new algorithm. We study a way to track faces using the neural oscillatory network which imitates the artificial neural net of information handing ability of human and animals, and biological movement characteristic of a singular neuron. The system that is suggested in this paper can broadly be broken into two stages of process. The first stage is the process of face extraction, which involves the acquisition of real-time RGB24bit color video delivering with the use of a cheap webcam. LEGION(Locally Excitatory Globally Inhibitory)algorithm is suggested as the face extraction method to be preceded for face tracking. The second stage is a method for face tracking by discovering the leader neuron that has the greatest connection strength amongst neighbor neuron of extracted face area. Along with the suggested method, the necessary element of face track such as stability as well as scale problem can be resolved.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.113-127
• /
• 2012

Face detection methods using image processing have been proposed in many different ways. Generally, the most widely used method for face detection is an Adaboost that is proposed by Viola and Jones. This method uses Haar-like feature for image learning, and the detection performance depends on the learned images. It is well performed to detect face images within a certain distance range, but if the image is far away from the camera, face images become so small that may not detect them with the pre-learned Haar-like feature of the face image. In this paper, we propose the far distance face detection method that combine the Aadaboost of Viola-Jones with a saliency map and user's attention information. Saliency Map is used to select the candidate face images in the input image, face images are finally detected among the candidated regions using the Adaboost with Haar-like feature learned in advance. And the user's eye-tracking information is used to select the interest regions. When a subject is so far away from the camera that it is difficult to detect the face image, we expand the small eye gaze spot region using linear interpolation method and reuse that as input image and can increase the face image detection performance. We confirmed the proposed model has better results than the conventional Adaboost in terms of face image detection performance and computational time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.4C
• /
• pp.476-483
• /
• 2009

Tracking and detecting targets by the microwave seeker is affected by the clutter reflecting from the earth's surface. In order to detect retreating targets in look-down scenario, which appear in the sidelobe clutter (SLC) region, in the microwave seeker of high pulse repetition frequency (HPRF) mode, it is necessary to understand statistical characteristics of the surface SLC. Statistical analysis of SLC has been conducted for several kinds of the surface using data obtained by the captive flight test of the microwave seeker in the HPRF mode. The probability density function (PDF) fitting is conducted for several kinds and conditions of the surface. PDFs and PDF parameters, which best describe statistical distribution of the SLC power, are estimated. By using the estimated PDFs and PDF parameters, analyses for setting the target-detection thresholds, which give a desired level of target-detection false alarm probability, are made. These analysis materials for statistical characteristics of SLC power and the target-detection threshold can be used in various fields, such as development of a target-detection method, the constant false alarm rate processing.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.298-301
• /
• 2007

Computational speed plays key role in background subtraction and shadow detection, because those are only preprocessing steps of a moving object segmentation, tracking and activity recognition. A color intensity variation based approach fastly detect a moving object and extract shadow in a image sequences. The moving object is subtracted from background using meanmax, meanmin thresholds and shadow is detected by decrease limit and correspondence thresholds. The proposed approach relies on the ability to represent shadow cast impact by offline experiment dataset on sub grouped RGB color space.

• Current Optics and Photonics
• /
• v.4 no.5
• /
• pp.421-427
• /
• 2020

A novel privacy-protection device to prevent visual hacking is realized by using a directional backlight and facial recognition. The proposed method is able to overcome the limitations of previous privacy-protection methods that simply restrict the viewing angle to a narrow range. The accuracy of user tracking is accomplished by the combination of a time-of-flight sensor and facial recognition with no restriction of detection range. In addition, an experimental demonstration is provided to verify the proposed scheme.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.25 no.4
• /
• pp.327-336
• /
• 2021

In this paper, we will introduce the image processing methods for the remote pupillary light reflex measurement using the video taken by a general smartphone camera without a special device such as an infrared camera. We propose an algorithm for estimate the size of the pupil that changes with light using image data analysis without a learning process. In addition, we will introduce the results of visualizing the change in the pupil size by removing noise from the recorded data of the pupil size measured for each frame of the video. We expect that this study will contribute to the construction of an objective indicator for remote pupillary light reflex measurement in the situation where non-face-to-face communication has become common due to COVID-19 and the demand for remote diagnosis is increasing.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.2
• /
• pp.23-29
• /
• 2014

The advanced traffic management system of intelligent transport systems automates the related traffic tasks such as vehicle speed, traffic volume and traffic incidents through the improved infrastructures like high definition cameras, high-performance radar sensors. For the safety of road users, especially, the automated incident detection and secondary accident prevention system is required. Normally, CCTV based image object detection and radar based object detection is used in this system. In this paper, we proposed the algorithm for real time highway incident detection system using multi surveillance cameras to mosaic video and track accurately the moving object that taken from different angles by background modeling. We confirmed through experiments that the video detection can supplement the short-range shaded area and the long-range detection limit of radar. In addition, the video detection has better classification features in daytime detection excluding the bad weather condition.

• Journal of Korean Society of Transportation
• /
• v.27 no.5
• /
• pp.97-111
• /
• 2009

Image-based traffic information collection systems have entered widespread adoption and use in many countries since these systems are not only capable of replacing existing loop-based detectors which have limitations in management and administration, but are also capable of providing and managing a wide variety of traffic related information. In addition, these systems are expanding rapidly in terms of purpose and scope of use. Currently, the utilization of image processing technology in the field of traffic accident management is limited to installing surveillance cameras on locations where traffic accidents are expected to occur and digitalizing of recorded data. Accurately recording the sequence of situations around a traffic accident in a signal intersection and then objectively and clearly analyzing how such accident occurred is more urgent and important than anything else in resolving a traffic accident. Therefore, in this research, we intend to present a technology capable of overcoming problems in which advanced existing technologies exhibited limitations in handling real-time due to large data capacity such as object separation of vehicles and tracking, which pose difficulties due to environmental diversities and changes at a signal intersection with complex traffic situations, as pointed out by many past researches while presenting and implementing an active and environmentally adaptive methodology capable of effectively reducing false detection situations which frequently occur even with the Gaussian complex model analytical method which has been considered the best among well-known environmental obstacle reduction methods. To prove that the technology developed by this research has performance advantage over existing automatic traffic accident recording systems, a test was performed by entering image data from an actually operating crossroad online in real-time. The test results were compared with the performance of other existing technologies.

• Journal of Sensor Science and Technology
• /
• v.1 no.2
• /
• pp.155-163
• /
• 1992

The phased array antenna has the ability to perform adaptive sampling by directing the radar beam without inertia in any direction. The adaptive sampling capability of the phased array antenna allows each sampling time interval to be varied for each target, depending on the acceleration of each target at any time. In this paper we design a three-dimensional adaptive tracking algorithm for the phased array radar system with a given set of measurement parameters. The tracking algorithm avoids taking unnecessarily frequent samples, while keeping the angular prediction error within a fraction of antenna beamwidth so that the probability of detection will not be degraded during a track update illuminations. In our algorithm, the target model and the sampling rate are selected depending on the target range and the target maneuver status which is determined by a maneuver detector. A detailed simulation is conducted to test the validity of our tracking algorithm for encounter geometries under various conditions of maneuver.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.12
• /
• pp.127-134
• /
• 2020

Recently, Harmful sites, including pornographic videos, drugs, personal information and hacking tool distribution sites, have caused serious social problems. However, due to the nature of the Internet environment where anyone can use it freely, it is difficult to control the user effectively. And the site operator operates by changing the domain to bypass the blockage. Therefore, even once identified sites have low persistence. In this paper, we propose multi-channel domain tracking technology, a technique that can effectively track changes in the domain addresses of harmful sites, including the same or similar content, by tracking changes in these harmful sites. Proposed technology is a technology that can continuously track information in a domain using OSINT technology. We tested and verified that the proposed technology was effective for domain tracking with a 90.4% trace rate (sensing 66 changes out of 73 domains).