• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.847-850
• /
• 2008

On today, human is going to develop machine by using mutual communication to machine. Including vision - based HCI(Human Computer Interaction), the technique which to recognize finger and to track finger is important in HCI systems, in HCI systems. In order to divide finger, this paper uses more effectively dividing the technique using subtraction which is separation of background and foreground, as well as to divide finger from limited background and cluttered background. In order to divide finger, the finger is recognized to make "Template-Matching" by identified fingertip images. And, identified gestures be compared the tracked gesture after tracking recognized finger. In this paper, after obtaining interest area, not only using subtraction image and template-matching but to perform template-matching in the area. So, emphasis is placed on decreasing perform speed and reaction speed, and we propose technique which is more effectively recognizing gestures.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.422-424
• /
• 2021

In this paper, it leads the approach of fusing multiple RGB cameras for visual objects recognition based on deep learning with convolution neural network and 3D Light Detection and Ranging (LiDAR) to observe the environment and match into a 3D world in estimating the distance and position in a form of point cloud map. The goal of perception in multiple cameras are to extract the crucial static and dynamic objects around the autonomous vehicle, especially the blind spot which assists the AV to navigate according to the goal. Numerous cameras with object detection might tend slow-going the computer process in real-time. The computer vision convolution neural network algorithm to use for eradicating this problem use must suitable also to the capacity of the hardware. The localization of classified detected objects comes from the bases of a 3D point cloud environment. But first, the LiDAR point cloud data undergo parsing, and the used algorithm is based on the 3D Euclidean clustering method which gives an accurate on localizing the objects. We evaluated the method using our dataset that comes from VLP-16 and multiple cameras and the results show the completion of the method and multi-sensor fusion strategy.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.382-384
• /
• 2021

In this study, we implement a system that automatically classifies types of plastics using YOLO (You Only Look Once), a real-time object recognition algorithm. The system consists of Nvidia jetson nano, a small computer for deep learning and computer vision, with model trained to recognize plastic separation emission marks using YOLO. Using a webcam, recycling marks of plastic waste were recognized as PET, HDPE, and PP, and motors were adjusted to be classified according to the type. By implementing this automatic classifier, it is convenient in that it can reduce the labor of separating and discharging plastic separation marks by humans and increase the efficiency of recycling through accurate recycling.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.11 no.2
• /
• pp.118-123
• /
• 2011

Video based object tracking normally deals with non-stationary image streams that change over time. Robust and real time moving object tracking is considered to be a problematic issue in computer vision. Multiple object tracking has many practical applications in scene analysis for automated surveillance. In this paper, we introduce a specified object tracking based particle filter used in an environment of multiple moving objects. A differential image region based tracking method for the detection of multiple moving objects is used. In order to ensure accurate object detection in an unconstrained environment, a background image update method is used. In addition, there exist problems in tracking a particular object through a video sequence, which cannot rely only on image processing techniques. For this, a probabilistic framework is used. Our proposed particle filter has been proved to be robust in dealing with nonlinear and non-Gaussian problems. The particle filter provides a robust object tracking framework under ambiguity conditions and greatly improves the estimation accuracy for complicated tracking problems.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.277-280
• /
• 2016

Real-time remote monitoring system has an important value in many surveillance situations. It allows someone to be informed of what is happening in his monitoring locations. Kinect v2 is a new kind of camera which gives computers eyes and can generate different data such as color and depth images, audio input and skeletal data. In this paper, using Kinect v2 sensor with its depth image, we present a monitoring system in a space covered by Kinect. Therefore, based on space covered by Kinect camera, we define a target area to monitor using depth range by setting minimum and maximum distances. With computer vision library (Emgu CV), if there is an object tracked in the target space, kinect camera captures the whole image color and sends it in database and user gets at the same time a notification on his mobile device wherever he is with internet access.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Journal of KIISE:Software and Applications
• /
• v.30 no.12
• /
• pp.1228-1238
• /
• 2003

This paper presents a new and practical method based on computer vision for detecting the monitor position where the user is looking. In general, the user tends to move both his face and eyes in order to gaze at certain monitor position. Previous researches use only one wide view camera, which can capture a whole user's face. In such a case, the image resolution is too low and the fine movements of user's eye cannot be exactly detected. So, we implement the gaze detection system with dual camera systems(a wide and a narrow view camera). In order to locate the user's eye position accurately, the narrow view camera has the functionalities of auto focusing and auto panning/tilting based on the detected 3D facial feature positions from the wide view camera. In addition, we use dual R-LED illuminators in order to detect facial features and especially eye features. As experimental results, we can implement the real-time gaze detection system and the gaze position accuracy between the computed positions and the real ones is about 3.44 cm of RMS error.

• Journal of Broadcast Engineering
• /
• v.21 no.4
• /
• pp.493-505
• /
• 2016

Among modern computer vision and video surveillance systems, the pedestrian counting system is a one of important systems in terms of security, scheduling and advertising. In the field of, pedestrian counting remains a variety of challenges such as changes in illumination, partial occlusion, overlap and people detection. During pedestrian counting process, the biggest problem is occlusion effect in crowded environment. Occlusion and overlap must be resolved for accurate people counting. In this paper, we propose a novel pedestrian counting system which improves existing pedestrian tracking method. Unlike existing pedestrian tracking method, proposed method shows that average filter tracking method can improve tracking performance. Also proposed method improves tracking performance through frame compensation and outlier removal. At the same time, we keep various information of tracking objects. The proposed method improves counting accuracy and reduces error rate about S6 dataset and S7 dataset. Also our system provides real time detection at the rate of 80 fps.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.6
• /
• pp.117-125
• /
• 2008

The Object tracking by real-time image analysis is one of the major concerns in computer vision and its application fields. The Object detection process of real-time images must be preceded before the object tracking process. To achieve the stable object detection performance in the exterior environment, adaptive background model generation methods are needed. The adaptive background model can accept the nature's phenomena changes and adapt the system to the changes such as light or shadow movements that are caused by changes of meridian altitudes of the sun. In this paper, we propose a robust background model generation method effective in an illegal parking auto-detection application area. We also provide a evaluation method that judges whether a moving vehicle stops or not. As the first step, an initial background model is generated. Then the differences between the initial model and the input image frame is used to trace the movement of object. The moving vehicle can be easily recognized from the object tracking process. After that, the model is updated by the background information except the moving object. These steps are repeated. The experiment results show that our background model is effective and adaptable in the variable exterior environment. The results also show our model can detect objects moving slowly. This paper includes the performance evaluation results of the proposed method on the real roads.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.279-282
• /
• 2017

Kinect depth sensor, a depth camera developed by Microsoft as a natural user interface for game appeared as a very useful tool in computer vision field. In this paper, due to kinect's depth sensor and its high frame rate, we developed a distance measurement system using Kinect camera to test it for unmanned vehicles which need vision systems to perceive the surrounding environment like human do in order to detect objects in their path. Therefore, kinect depth sensor is used to detect objects in its field of view and enhance the distance measurement system from objects to the vision sensor. Detected object is identified in accuracy way to determine if it is a real object or a pixel nose to reduce the processing time by ignoring pixels which are not a part of a real object. Using depth segmentation techniques along with Open CV library for image processing, we can identify present objects within Kinect camera's field of view and measure the distance from them to the sensor. Tests show promising results that this system can be used as well for autonomous vehicles equipped with low-cost range sensor, Kinect camera, for further processing depending on the application type when they reach a certain distance far from detected objects.