• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.5
• /
• pp.2430-2436
• /
• 2013

This study suggests a method of detecting the still object, which becomes a cause of danger within the crossroad. The Inverse Perspective Transform was performed in order to make the object size consistent by being inputted the real-time image from CCTV that is installed within the crossroad. It established the detection area in the image with the perspective transform and generated the adaptative background image with the use of the moving information on object. The detection of the stop object was detected the candidate region of the stop object by using the background-image differential method. To grasp the appearance of truth on the detected candidate region, a method is proposed that uses the gradient information on image and EHD(Edge Histogram Descriptor). To examine performance of the suggested algorithm, it experimented by storing the images in the commuting time and the daytime through DVR, which is installed on the cross street. As a result of experiment, it could efficiently detect the stop vehicle within the detection region inside the crossroad. The processing speed is shown in 13~18 frame per second according to the area of the detection region, thereby being judged to likely have no problem about the real-time processing.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.3
• /
• pp.70-77
• /
• 2017

Recently, Internet of Things (IoT) and deep learning techniques have affected video surveillance systems in various ways. The surveillance features that perform detection, tracking, and classification of specific objects in Closed Circuit Television (CCTV) video are becoming more intelligent. This paper presents real-time algorithm that can run in a PC environment using only a low power CPU. Traditional tracking algorithms combine background modeling using the Gaussian Mixture Model (GMM), Hungarian algorithm, and a Kalman filter; they have relatively low complexity but high detection errors. To supplement this, deep learning technology was used, which can be trained from a large amounts of data. In particular, an SRGB(Sequential RGB)-3 Layer CNN was used on tracked objects to emphasize the features of moving people. Performance evaluation comparing the proposed algorithm with existing ones using HOG and SVM showed move-in and move-out error rate reductions by 7.6 % and 9.0 %, respectively.

• 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.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.4
• /
• pp.641-647
• /
• 2018

Recently, CCTV, which provides video information for multiple purposes, has been transformed into an intelligent, and the range of automation applications increased using the computer vision. A highly reliable detection method must be performed for accurate recognition of pedestrians and vehicles and various methods are being studied for this purpose. In such an object detection system. In this paper, we propose a method to detect a large number of pedestrians by acquiring three characteristic information that features of color information using HSI, motion vector information and shaping information using HOG feature information of a pedestrian in a situation where a large number of pedestrians are moving. The proposed method distinguishes each pedestrian while minimizing the failure or confusion of pedestrian detection and tracking. Also when pedestrians approach or overlap, pedestrians are identified and detected using stored frame feature data.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.5
• /
• pp.189-193
• /
• 2024

According to the Korean Road Traffic Act, continuous driving in the overtaking lane (1 lane) of the highway is judged as a violation of the designated lane. Currently, in order to crack down on the advanced situation on the highway, a citizen's report or the road police directly determine whether it is a violation and crack down. This is because a violation is judged by overtaking or not the speed of the vehicle on the highway, and it is difficult to judge whether the vehicle is continuously driving because the standard is ambiguous in CCTV. Therefore, a system that self-determines and regulates whether the first lane is continuously driving without human intervention is needed. In this paper, in order to enable multiple object tracking during object tracking and to ensure the system's real-time feasibility, an unmanned crackdown system was designed based on F-SORT (Focused-Simple Online and Realtime Tracking) based on the Simple Online and Realtime Tracking (SORT) model, and the system determines whether or not the vehicle is continuously driving in one lane by determining the moving distance of the vehicle

• Journal of Internet Computing and Services
• /
• v.17 no.6
• /
• pp.1-8
• /
• 2016

Current indoor intrusion detection and location tracking methods have the weakness in seamless operations in tracking the objective because the object must possess a communicating device and the limitation of the single cell size (approximate $100cm{\times}100cm$) exits. Also, the utilization of CCTV technologies show the shortcomings in tracking when the object disappear the area where the CCTV is not installed or illumination is not enough for capturing the scene (e.g. where the context-awarded system is not installed or low illumination presents). Therefore, in this paper we present an improved in-door tracking system based on sensor networks. Such system is built on a simulated scenario and enables us to detect and extend the area of surveillance as well as actively responding the emergency situation. Through simulated studies, we have demonstrated that the proposed system is capable of supplementing the shortcomings of signal cutting, and of estimating the location of the moving object. We expect the study will improve the better analysis of the intruder behavior, the more effective prevention and flexible response to various emergency situations.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.20 no.6
• /
• pp.1161-1175
• /
• 2018

An unexpected event could be easily followed by a large secondary accident due to the limitation in sight of drivers in road tunnels. Therefore, a series of automated incident detection systems have been under operation, which, however, appear in very low detection rates due to very low image qualities on CCTVs in tunnels. In order to overcome that limit, deep learning based tunnel incident detection system was developed, which already showed high detection rates in November of 2017. However, since the object detection process could deal with only still images, moving direction and speed of moving vehicles could not be identified. Furthermore it was hard to detect stopping and reverse the status of moving vehicles. Therefore, apart from the object detection, an object tracking method has been introduced and combined with the detection algorithm to track the moving vehicles. Also, stopping-reverse discrimination algorithm was proposed, thereby implementing into the combined incident detection processes. Each performance on detection of stopping, reverse driving and fire incident state were evaluated with showing 100% detection rate. But the detection for 'person' object appears relatively low success rate to 78.5%. Nevertheless, it is believed that the enlarged richness of image big-data could dramatically enhance the detection capacity of the automatic incident detection system.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.418-427
• /
• 2016

A smart parking management requires to track a vehicle in a indoor parking lot and to detect the place where the vehicle is parked. An advanced parking system watches all space of the parking lot with CCTV cameras. We can use these cameras for vehicles tracking and detection. In order to cover a wide area with a camera, a fisheye lens is used. In this case the shape and size of an moving vehicle vary much with distance and angle to the camera. This makes vehicle detection and tracking difficult. In addition to the fisheye lens, the vehicle headlights also makes vehicle detection and tracking difficult. This paper describes a method of realtime vehicle detection and tracking robust to the harsh situation described above. In each image frame, we update the region of a vehicle and estimate the vehicle movement. First we approximate the shape of a car with a quadrangle and estimate the four sides of the car using multiple histograms of oriented gradient. Second we create a template by applying a distance transform to the car region and estimate the motion of the car with a template matching method.

• Annual Conference of KIPS
• /
• 2017.04a
• /
• pp.833-836
• /
• 2017

최근 블랙박스 및 CCTV 같은 영상 촬영 장치가 보편화되면서, 방대한 양의 영상 데이터가 실시간으로 생성되고 있다. 만약 이 대용량 데이터 안의 차량 정보를 추출할 수 있다면 범죄 차량 추적, 교통 혼잡도 측정 등의 활용이 가능할 것이다. 이를 구현하기 위해서는 수많은 자동차에서 실시간으로 생성되는 영상 데이터를 처리할 수 있는 시스템이 필수적이나, 이러한 시스템을 찾기 힘든 것이 현실이다. 이를 위해 이 논문에서는 아파치 카프카, Hbase를 이용한 영상 빅데이터 처리 시스템을 제안한다. 아파치 카프카는 시스템 내에서 영상 손실이 없는 전송과 영상 처리 노드의 스케줄링을 수행하며, Hbase는 처리된 데이터를 테이블로 저장하고 사용자가 보낸 쿼리를 처리한다. 더불어, Hbase에 인덱스를 구성하여 빠른 쿼리 처리가 가능하도록 만든다. 실험 결과, 제안된 시스템은 인덱스가 없을 때보다 뛰어난 쿼리 처리 성능을 보이는 것을 확인할 수 있었다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.45-55
• /
• 2021

This paper has proposed an algorithm that detecting for dense small vehicle in large image efficiently. It is consisted of two Ensemble Deep-Learning Network algorithms based on Coarse to Fine method. The system can detect vehicle exactly on selected sub image. In the Coarse step, it can make Voting Space using the result of various Deep-Learning Network individually. To select sub-region, it makes Voting Map by to combine each Voting Space. In the Fine step, the sub-region selected in the Coarse step is transferred to final Deep-Learning Network. The sub-region can be defined by using dynamic windows. In this paper, pre-defined mapping table has used to define dynamic windows for perspective road image. Identity judgment of vehicle moving on each sub-region is determined by closest center point of bottom of the detected vehicle's box information. And it is tracked by vehicle's box information on the continuous images. The proposed algorithm has evaluated for performance of detection and cost in real time using day and night images captured by CCTV on the road.