• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.634-636
• /
• 2023

Surveillance systems play a pivotal role in ensuring the safety and security of various environments, including public spaces, critical infrastructure, and private properties. However, detecting abnormal human behavior in lowlight conditions is a critical yet challenging task due to the inherent limitations of visual data acquisition in such scenarios. This paper introduces a spatiotemporal framework designed to address the unique challenges posed by low-light environments, enhancing the accuracy and efficiency of human abnormality detection in surveillance camera systems. We proposed the pre-processing using lightweight exposure correction, patched frames pose estimation, and optical flow to extract the human behavior flow through t-seconds of frames. After that, we train the estimated-action-flow into autoencoder for abnormal behavior classification to get normal loss as metrics decision for normal/abnormal behavior.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1448-1451
• /
• 2004

Position estimation is one of the most important functions for the mobile robot navigating in the unstructured environment. Most of previous localization schemes estimate current position and pose of mobile robot by applying various localization algorithms with the information obtained from sensors which are set on the mobile robot, or by recognizing an artificial landmark attached on the wall, or objects of the environment as natural landmark in the indoor environment. Several drawbacks about them have been brought up. To compensate the drawbacks, a new localization method that estimates the absolute position of the mobile robot by using a fixed camera on the ceiling in the corridor is proposed. And also, it can improve the success rate for position estimation using the proposed method, which calculates the real size of an object. This scheme is not a relative localization, which decreases the position error through algorithms with noisy sensor data, but a kind of absolute localization. The effectiveness of the proposed localization scheme is demonstrated through the experiments.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1934-1935
• /
• 2011

In order for a service robot to perform several tasks, basically autonomous navigation technique such as localization, mapping, and path planning is required. The localization (estimation robot's pose) is fundamental ability for service robot to navigate autonomously. In this paper, we propose a new system for point pattern matching based visual global localization using spot lightings in ceiling. The proposed algorithm us suitable for system that demands high accuracy and fast update rate such a guide robot in the exhibition. A single camera looking upward direction (called ceiling vision system) is mounted on the head of the mobile robot and image features such as lightings are detected and tracked through the image sequence. For detecting more spot lightings, we choose wide FOV lens, and inevitably there is serious image distortion. But by applying correction calculation only for the position of spot lightings not whole image pixels, we can decrease the processing time. And then using point pattern matching and least square estimation, finally we can get the precise position and orientation of the mobile robot. Experimental results demonstrate the accuracy and update rate of the proposed algorithm in real environments.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.2
• /
• pp.84-91
• /
• 2009

This paper presents an online partial 3D modeling methodology that uses a mobile augmented reality system and aerial photographs, and a tracking methodology that compares the 3D model with a video image. Instead of relying on models which are created in advance, the system generates a 3D model for a real building on the fly by combining frontal and aerial views. A user's initial pose is estimated using an aerial photograph, which is retrieved from a database according to the user's GPS coordinates, and an inertial sensor which measures pitch. We detect edges of the rooftop based on Graph cut, and find edges and a corner of the bottom by minimizing the proposed cost function. To track the user's position and orientation in real-time, feature-based tracking is carried out based on salient points on the edges and the sides of a building the user is keeping in view. We implemented camera pose estimators using both a least squares estimator and an unscented Kalman filter (UKF). We evaluated the speed and accuracy of both approaches, and we demonstrated the usefulness of our computations as important building blocks for an Anywhere Augmentation scenario.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.127-133
• /
• 2021

A recent study of people physically falling focused on analyzing the motions of the falls using a recurrent neural network (RNN) and a deep learning approach to get good results from detecting 2D human poses from a single color image. In this paper, we investigate a detection method for estimating the position of the head and shoulder keypoints and the acceleration of positional change using the skeletal keypoints information extracted using PoseNet from an image obtained with a low-cost 2D RGB camera, increasing the accuracy of judgments about the falls. In particular, we propose a fall detection method based on the characteristics of post-fall posture in the fall motion-analysis method. A public data set was used to extract human skeletal features, and as a result of an experiment to find a feature extraction method that can achieve high classification accuracy, the proposed method showed a 99.8% success rate in detecting falls more effectively than a conventional, primitive skeletal data-use method.

• Journal of Korea Multimedia Society
• /
• v.10 no.3
• /
• pp.303-315
• /
• 2007

This paper proposes a 3D structure recovery and registration method that uses four or more common points. For each frame of a given video, a partial structure is recovered using tracked points. The 3D coordinates, camera positions and camera directions are computed at once by our improved scaled orthographic factorization method. The partially recovered point sets are parts of a whole model. A registration of point sets makes the complete shape. The recovered subsets are integrated by transforming each coordinate system of the local point subset into a common basis coordinate system. The process of shape recovery and integration is performed uniformly and linearly without any nonlinear iterative process and without loss of accuracy. The execution time for the integration is significantly reduced relative to the conventional ICP method. Due to the fast recovery and registration framework, our shape recovery scheme is applicable to various interactive video applications. The processing time per frame is under 0.01 seconds in most cases and the integration error is under 0.1mm on average.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.4
• /
• pp.349-355
• /
• 2015

Underwater robots generally show better performances for tasks than humans under certain underwater constraints such as. high pressure, limited light, etc. To properly diagnose in an underwater environment using remotely operated underwater vehicles, it is important to keep autonomously its own position and orientation in order to avoid additional control efforts. In this paper, we propose an efficient method to assist in the operation for the various disturbances of a remotely operated vehicle for the diagnosis of underwater structures. The conventional AHRS-based bearing estimation system did not work well due to incorrect measurements caused by the hard-iron effect when the robot is approaching a ferromagnetic structure. To overcome this drawback, we propose a sensor fusion algorithm with the camera and AHRS for estimating the pose of the ROV. However, the image information in the underwater environment is often unreliable and blurred by turbidity or suspended solids. Thus, we suggest an efficient method for fusing the vision sensor and the AHRS with a criterion which is the amount of blur in the image. To evaluate the amount of blur, we adopt two methods: one is the quantification of high frequency components using the power spectrum density analysis of 2D discrete Fourier transformed image, and the other is identifying the blur parameter based on cepstrum analysis. We evaluate the performance of the robustness of the visual odometry and blur estimation methods according to the change of light and distance. We verify that the blur estimation method based on cepstrum analysis shows a better performance through the experiments.

• Journal of Korea Multimedia Society
• /
• v.11 no.11
• /
• pp.1575-1591
• /
• 2008

In this paper, we propose a new method detecting foot motion capture in order to overlap in realtime foot's 3D virtual model from stereo cameras. In order to overlap foot's virtual model at the same position of the foot, a process of the foot's joint detection to regularly track the foot's joint motion is necessary, and accurate register both foot's virtual model and user's foot in complicated motion is most important problem in this technology. In this paper, we propose a dynamic registration using two types of marker groups. A plane information of the ground handles the relationship between foot's virtual model and user's foot and obtains foot's pose and location. Foot's rotation is predicted by two attached marker groups according to instep of center framework. Consequently, we had implemented our proposed system and estimated the accuracy of the proposed method using various experiments.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.7
• /
• pp.279-286
• /
• 2021

Recently, We are carrying out a policy of physical distancing of at least 1m from each other to prevent the spreading of COVID-19 disease in public places. In this paper, we propose a method for measuring distances between people in real time and an automation system that recognizes objects that are within 1 meter of each other from stereo images acquired by drones or CCTVs according to the estimated distance. A problem with existing methods used to estimate distances between multiple objects is that they do not obtain three-dimensional information of objects using only one CCTV. his is because three-dimensional information is necessary to measure distances between people when they are right next to each other or overlap in two dimensional image. Furthermore, they use only the Bounding Box information to obtain the exact coordinates of human existence. Therefore, in this paper, to obtain the exact two-dimensional coordinate value in which a person exists, we extract a person's key point to detect the location, convert it to a three-dimensional coordinate value using Stereo Vision and Camera Calibration, and estimate the Euclidean distance between people. As a result of performing an experiment for estimating the accuracy of 3D coordinates and the distance between objects (persons), the average error within 0.098m was shown in the estimation of the distance between multiple people within 1m.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.3
• /
• pp.1189-1204
• /
• 2018

Detecting and capturing 3D human structures from the intensity-based image sequences is an inherently arguable problem, which attracted attention of several researchers especially in real-time activity recognition (Real-AR). These Real-AR systems have been significantly enhanced by using depth intensity sensors that gives maximum information, in spite of the fact that conventional Real-AR systems are using RGB video sensors. This study proposed a depth-based routine-logging Real-AR system to identify the daily human activity routines and to make these surroundings an intelligent living space. Our real-time routine-logging Real-AR system is categorized into two categories. The data collection with the use of a depth camera, feature extraction based on joint information and training/recognition of each activity. In-addition, the recognition mechanism locates, and pinpoints the learned activities and induces routine-logs. The evaluation applied on the depth datasets (self-annotated and MSRAction3D datasets) demonstrated that proposed system can achieve better recognition rates and robust as compare to state-of-the-art methods. Our Real-AR should be feasibly accessible and permanently used in behavior monitoring applications, humanoid-robot systems and e-medical therapy systems.