• ETRI Journal
• /
• 제46권4호
• /
• pp.671-682
• /
• 2024

Violence can be committed anywhere, even in crowded places. It is hence necessary to monitor human activities for public safety. Surveillance cameras can monitor surrounding activities but require human assistance to continuously monitor every incident. Automatic violence detection is needed for early warning and fast response. However, such automation is still challenging because of low video resolution and blind spots. This paper uses ResNet50v2 and the gated recurrent unit (GRU) algorithm to detect violence in the Movies, Hockey, and Crowd video datasets. Spatial features were extracted from each frame sequence of the video using a pretrained model from ResNet50V2, which was then classified using the optimal trained model on the GRU architecture. The experimental results were then compared with wavelet feature extraction methods and classification models, such as the convolutional neural network and long short-term memory. The results show that the proposed combination of ResNet50V2 and GRU is robust and delivers the best performance in terms of accuracy, recall, precision, and F1-score. The use of ResNet50V2 for feature extraction can improve model performance.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2011년도 정기총회 및 추계학술대회 논문집
• /
• pp.768-773
• /
• 2011

Today's urban transit system is regarded as the important public transportation service which saves passengers' time and provides the safety. Many researches focus on the rapid and protective responses that minimize the losses when dangerous situation occurs. In this paper we proposed the early fire detection and corresponding rapid response method in urban transit system by combining automatic fire detection for video input and the sensor system. The fire detection method consists of two parts, spark detection and smoke detection. At the spark detection, the RGB color of input video is converted into HSV color and the frame difference is obtained in temporal direction. The region with high R values is considered as fire region candidate and stepwise fire detection rule is applied to calculate its size. At the smoke detection stage, we used the smoke sensor network to secure the credibility of spark detection. The proposed system can be implemented at low prices. In the future work, we would improve the detection algorithm and the accuracy of sensor location in the network.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권11호
• /
• pp.4869-4873
• /
• 2016

Colonoscopy is currently the best technique available for the detection of colon cancer or colorectal polyps or other precursor lesions. Computer aided detection (CAD) is based on very complex pattern recognition. Local binary patterns (LBPs) are strong illumination invariant texture primitives. Histograms of binary patterns computed across regions are used to describe textures. Every pixel is contrasted relative to gray levels of neighbourhood pixels. In this study, colorectal polyp detection was performed with colonoscopy video frames, with classification via J48 and Fuzzy. Features such as color, discrete cosine transform (DCT) and LBP were used in confirming the superiority of the proposed method in colorectal polyp detection. The performance was better than with other current methods.

• Journal of Multimedia Information System
• /
• 제5권3호
• /
• pp.189-194
• /
• 2018

In this paper, we propose the drowsiness detection method for car driver. This paper determines whether or not the driver's eyes are closed using the depth and infrared videos. The proposed method has the advantage to detect drowsiness without being affected by illumination. The proposed method detects a face in the depth picture by using the fact that the nose is closest to the camera. The driver's eyes are detected by using the extraction of harr-like feature within the detected face region. This method considers to be drowsiness if eyes are closed for a certain period of time. Simulation results show the drowsiness detection performance for the proposed method.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.95.2-95
• /
• 2002

1. Introduction of Camera Detection system Camera Detection system is an equipment that can detect realtime traffic information by image processing techniques. This information can be used to analyze and control road traffic flow. It is also used as a method to detect and control traffic flow for ITS(Intelligent Transportation System). Traffic information includes speed, head way, traffic flow, occupation time and length of queue. There are many detection systems for traffic data. But video detection system can detect multiple lanes with only one camera and collect various traffic information. So it is thought to be the most efficient method of all detection system. Though the...

• 한국항행학회논문지
• /
• 제13권6호
• /
• pp.1011-1016
• /
• 2009

본 논문에서는 공공장소에 광범위하게 설치되어있는 CCTV의 감시 기능을 활용하여 화재 발생 감지 방법을 제안하였다. 제안한 방법은 고정된 카메라로부터 칼라 정보를 이용하여 비디오 시퀀스의 화재 프레임 후보를 찾아내고, 공간 기법을 기반으로 감지된 화재 정보의 전경 색상을 분석하였다. 실험 결과, 비디오 시컨스에서 시 공간적 화재 후보 정보들이 급격히 변화할 때, 화재 감지의 성능이 우수함을 확인할 수 있었다.

• 반도체디스플레이기술학회지
• /
• 제20권3호
• /
• pp.117-119
• /
• 2021

With the development of the industry, the use of drones in specific mission flight is being actively studied. These drones fly a specified path and perform repetitive tasks. if the drone system will detect objects in real time, the performance of these mission flight will increase. In this paper, we implement object detection system and mount GPU acceleration to maximize the efficiency of limited device resources with drone video using Tensorflow Lite which enables in-device inference from a mobile device and Mobile SDK of DJI, a drone manufacture. For performance comparison, the average processing time per frame was measured when object detection was performed using only the CPU and when object detection was performed using the CPU and GPU at the same time.

• Journal of Information Processing Systems
• /
• 제15권5호
• /
• pp.1218-1230
• /
• 2019

This study proposes a novel video traffic flow detection method based on machine vision technology. The three-frame difference method, which is one kind of a motion evaluation method, is used to establish initial background image, and then a statistical scoring strategy is chosen to update background image in real time. Finally, the background difference method is used for detecting the moving objects. Meanwhile, a simple but effective shadow elimination method is introduced to improve the accuracy of the detection for moving objects. Furthermore, the study also proposes a vehicle matching and tracking strategy by combining characteristics, such as vehicle's location information, color information and fractal dimension information. Experimental results show that this detection method could quickly and effectively detect various traffic flow parameters, laying a solid foundation for enhancing the degree of automation for traffic management.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권2호
• /
• pp.169-177
• /
• 2009

본 논문에서는 뉴스 기사 분할 관점에서, 뉴스 비디오 샷 경계 탐지 알고리즘의 특성을 고려한 다음과 같은 설계 기준을 제시하고, 이를 모두 만족하는 새로운 샷 경계 탐지 알고리즘을 제안하고자 한다. 1) 뉴스 비디오 샷 경계 탐지의 재현율을 높임으로써, 앵커 샷 탐지 단계에서 입력으로 사용될 데이타의 오류를 최소화한다; 2) 급격한 장면 변환과 점증적 장면 변환을 하나의 알고리즘으로 탐지함으로써 한번의 데이타 탐색으로 샷 분할을 수행한다; 3) 분할된 샷들을 정적 샷과 동적 샷으로 분류함으로써 앵커샷 탐지 단계의 탐색 공간을 축소한다. 제안된 뉴스 비디오 샷 경계 탐지 알고리즘은 특이간 분해를 기반으로 점증적 클러스터링 알고리즘과 머서 커널을 결합한 구조로서, 위에서 제시한 기준을 모두 만족하도록 설계되었다. 제안된 방법론은 특이간 분해를 통해 특징 벡터의 차원축소 뿐만 아니라, 뉴스 비디오를 구성하는 연속적인 프레임에서의 잡음과 아주 작은 변화를 제거함으로써 분류 성능을 높일 수 있다. 또한 머서 커널의 도입은 쉽게 분류되지 않는 데이타를 고차원 공간으로 매핑함으로써 구분하기 어려운 샷 경계의 탐지 가능성을 높여준다. 실험을 통하여 제안된 방법론이 매우 높은 재현율을 보이며, 앵커 샷 탐지를 위한 탐색 공간 축소를 효과적으로 수행함을 보인다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.77-78
• /
• 2008

With advent of 3G mobile communication services, video telephony becomes one of the major services. However, due to a narrow channel bandwidth, the current video telephony services have not yet reached a satisfied level. In this paper, we propose an ROI (Region-Of-Interest) based improvement of visual quality for video telephony services with the H.264|MPEG-4 Part 10 (AVC: Advanced Video Coding) codec. To this end, we propose a face detection and tracking method to define ROI for the AVC codec based video telephony. Experiment results show that our proposed ROI based method allowed for improved visual quality in both objective and subjective perspectives.