• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.807-818
• /
• 2022

In order to further improve the accuracy and time efficiency of behavior recognition in intelligent monitoring scenarios, a human behavior recognition algorithm based on YOLO combined with LSTM and CNN is proposed. Using the real-time nature of YOLO target detection, firstly, the specific behavior in the surveillance video is detected in real time, and the depth feature extraction is performed after obtaining the target size, location and other information; Then, remove noise data from irrelevant areas in the image; Finally, combined with LSTM modeling and processing time series, the final behavior discrimination is made for the behavior action sequence in the surveillance video. Experiments in the MSR and KTH datasets show that the average recognition rate of each behavior reaches 98.42% and 96.6%, and the average recognition speed reaches 210ms and 220ms. The method in this paper has a good effect on the intelligence behavior recognition.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.1
• /
• pp.69-76
• /
• 2010

The segmentation of moving object in video sequence is a core technique of intelligent image processing system such as video surveillance, traffic monitoring and human tracking. A typical method to segment a moving region from the background is the background subtraction. The steps of background subtraction involve calculating a reference image, subtracting new frame from reference image and then thresholding the subtracted result. One of famous background modeling is Gaussian mixture model (GMM). Even though the method is known efficient and exact, GMM suffers from a problem that includes false pixels in ROI (region of interest), specifically shadow pixels. These false pixels cause fail of the post-processing tasks such as tracking and object recognition. This paper presents a method for removing false pixels included in ROT. First, we subdivide a ROI by using shape characteristics of detected objects. Then, a method is proposed to classify pixels from using histogram characteristic and comparing difference of energy that converts the color value of pixel into grayscale value, in order to estimate whether the pixels belong to moving object area or shadow area. The method is applied to real video sequence and the performance is verified.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1361-1370
• /
• 2016

To reduce huge losses in pig farms, weaning pigs with weak immune systems are required to be carefully supervised. Even if various researches have been performed for pig monitoring environment, segmenting each pig from touching-pigs is still entrenched as a difficult problem. In this paper, we propose a segmentation method for touching-pigs by using concave-points and edge information in a video surveillance system. Especially, we interpret the segmentation problem as a time-series analysis problem in order to identify the concave-points generated by touching-pigs. Based on the experimental results with the videos obtained from a domestic pig farm, we believe that the proposed method can accurately segment the touching-pigs.

• Fire Science and Engineering
• /
• v.32 no.6
• /
• pp.40-45
• /
• 2018

The omnidirectional surveillance camera uses the object detection algorithm to level the object by unit so that broadband surveillance can be performed using a fisheye lens and then, it was a field experiment with a system composed of an omnidirectional surveillance camera and a tracking (PTZ) camera. The omnidirectional surveillance camera accurately detects the moving object, displays the squarely, and tracks it in close cooperation with the tracking camera. In the field test of flame detection and temperature of the sensing camera, when the flame is detected during the auto scan, the detection camera stops and the temperature is displayed by moving the corresponding spot part to the central part of the screen. It is also possible to measure the distance of the flame from the distance of 1.5 km, which exceeds the standard of calorific value of 1 km 2,340 kcal. In the performance test of detecting the flame along the distance, it is possible to be 1.5 km in width exceeding $56cm{\times}90cm$ at a distance of 1km, and so it is also adaptable to forest fire. The system is expected to be very useful for safety such as prevention of intrinsic or surrounding fire and intrusion monitoring if it is installed in a petroleum gas storage facility or a storing place for oil in the future.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.4
• /
• pp.403-411
• /
• 2010

This research demonstrates a 3D virtual visualization of urban environment and video fusion for effective damage prevention and surveillance system using open standard. We present the visualization and interaction simulation method to increase the situational awareness and optimize the realization of environmental monitoring through the CCTV video and 3D virtual environment. New camera prototype was designed based on the camera frustum view model to project recorded video prospectively onto the virtual 3D environment. The demonstration was developed by the X3D, which is royalty-free open standard and run-time architecture, and it offers abilities to represent, control and share 3D spatial information via the internet browsers.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.10
• /
• pp.473-478
• /
• 2015

Recently, automatic video monitoring of individual pigs is emerging as an important issue in the management of group-housed pigs. Although a rich variety of studies have been reported on video monitoring techniques in intensive pig farming, it still requires further elaboration. In particular, when there exist adjoining pigs in a crowd pig room, it is necessary to have a way of separating adjoining pigs from the perspective of an image processing technique. In this paper, we propose an efficient image segmentation solution using both spatio-temporal information and region growing method for the identification of individual pigs in video surveillance systems. The experimental results with the videos obtained from a pig farm located in Sejong illustrated the efficiency of the proposed method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.5
• /
• pp.505-510
• /
• 2009

In recent years the space where a robot works has been expanding to the human space unlike traditional industrial robots that work only at fixed positions apart from humans. A human in the recent situation may be the owner of a robot or the target in a robotic application. This paper deals with the latter case; when a robot vision system is employed to monitor humans for a surveillance application, each person in a scene needs to be identified. Humans, however, often move together, and occlusions between them occur frequently. Although this problem has not been seriously tackled in relevant literature, it brings difficulty into later image analysis steps such as tracking and scene understanding. In this paper, a probabilistic neural network is employed to learn the patterns of the best dividing position along the top pixels of an image region of partly occlude people. As this method uses only shape information from an image, it is simple and can be implemented in real time.

• Journal of Digital Contents Society
• /
• v.18 no.6
• /
• pp.1009-1015
• /
• 2017

Recently, video surveillance is becoming more and more common as many camera are installed due to crime, terrorism, traffic and security. And systems that control cameras are becoming increasingly general. Video input from the installed camera is monitored by the multiscreen at the central control center, it is essential to simultaneously monitor multiscreen in real-time to quickly respond to situations or dangers. However, monitoring of multiscreen in a mobile environment such as a smart phone is not applied to hardware specifications or network bandwidth problems. For resolving these problems, in this paper, we propose a system that can monitor multiscreen in real-time in mobile-phone environment. We reconstruct the desired multiscreen through transcoding, it is possible to monitor continuously video streaming of multiple cameras, and to have the advantage of being mobile in mobile-phone environment.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.10
• /
• pp.3475-3489
• /
• 2014

Here we present a simple flame detection method for an infrared (IR) thermal camera based real-time fire surveillance digital signal processor (DSP) system. Infrared thermal cameras are especially advantageous for unattended fire surveillance. All-weather monitoring is possible, regardless of illumination and climate conditions, and the data quantity to be processed is one-third that of color videos. Conventional IR camera-based fire detection methods used mainly pixel-based temporal correlation functions. In the temporal correlation function-based methods, temporal changes in pixel intensity generated by the irregular motion and spreading of the flame pixels are measured using correlation functions. The correlation values of non-flame regions are uniform, but the flame regions have irregular temporal correlation values. To satisfy the requirement of early detection, all fire detection techniques should be practically applied within a very short period of time. The conventional pixel-based correlation function is computationally intensive. In this paper, we propose an IR camera-based simple flame detection algorithm optimized with a compact embedded DSP system to achieve early detection. To reduce the computational load, block-based calculations are used to select the candidate flame region and measure the temporal motion of flames. These functions are used together to obtain the early flame detection algorithm. The proposed simple algorithm was tested to verify the required function and performance in real-time using IR test videos and a real-time DSP system. The findings indicated that the system detected the flames within 5 to 20 seconds, and had a correct flame detection ratio of 100% with an acceptable false detection ratio in video sequence level.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2483-2503
• /
• 2016

Traffic congestion is a severe problem in many modern cities around the world. Real-time and accurate traffic congestion identification can provide the advanced traffic management systems with a reliable basis to take measurements. The most used data sources for traffic congestion are loop detector, GPS data, and video surveillance. Video based traffic monitoring systems have gained much attention due to their enormous advantages, such as low cost, flexibility to redesign the system and providing a rich information source for human understanding. In general, most existing video based systems for monitoring road traffic rely on stationary cameras and multiple vehicle tracking method. However, most commonly used multiple vehicle tracking methods are lack of effective track initiation schemes. Based on the motion of the vehicle usually obeys constant velocity model, a novel vehicle recognition method is proposed. The state of recognized vehicle is sent to the GM-PHD filter as birth target. In this way, we relieve the insensitive of GM-PHD filter for new entering vehicle. Combining with the advanced vehicle detection and data association techniques, this multiple vehicle tracking method is used to identify traffic congestion. It can be implemented in real-time with high accuracy and robustness. The advantages of our proposed method are validated on four real traffic data.