• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.8
• /
• pp.766-772
• /
• 2015

This paper proposes an object detection method based on motion estimation using background subtraction in the fisheye images obtained through omni-directional camera mounted on the vehicle. Recently, most of the vehicles installed with rear camera as a standard option, as well as various camera systems for safety. However, differently from the conventional object detection using the image obtained from the camera, the embedded system installed in the vehicle is difficult to apply a complicated algorithm because of its inherent low processing performance. In general, the embedded system needs system-dependent algorithm because it has lower processing performance than the computer. In this paper, the location of object is estimated from the information of object's motion obtained by applying a background subtraction method which compares the previous frames with the current ones. The real-time detection performance of the proposed method for object detection is verified experimentally on embedded board by comparing the proposed algorithm with the object detection based on LKOF (Lucas-Kanade optical flow).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.2
• /
• pp.285-293
• /
• 2017

Jump rope has two motions. It starts as hand motion and ends as jump motion. Therefore, two motions should be considered together to detect rotations accurately. But previous researches only consider one of the two motions as in push-up, sit-up, lift dumbbells etc, which results in inaccurate detection of rotations. In this paper, detection of rotation in jump rope using two motions through 6-axis accelerometer gyro sensor is proposed. Jump motion is detected using accelerometer sensor and hand motion is detected using gyro sensor. Also start point and end point of jump rope is detected using magnitude and standard deviation of accelerometer and gyro sensor values. The count of rotation is detected using y-axis of gyro sensor value. Y-axis of gyro sensor value indicate hand motion of jump rope motion. The usefulness of the proposed method is confirmed through experimental results.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.45-55
• /
• 2021

Information obtained through various sensors embedded in a smartphone can be used to identify and analyze user's movements and situations. In this paper, we propose two rule-based motion detection systems that can detect three alphabet motions, 'I', 'S', and 'Z' by analyzing data obtained by the acceleration and gyroscope sensors in a smartphone. First of all, the characteristics of acceleration and angular velocity for each motion are analyzed. Based on the analysis, two rule-based systems are proposed and implemented as an android application and it is used to verify the detection performance for each motion. Two rule-based systems show high recognition rate over 90% for each motion and the rule-based system using ensemble shows better performance than another one.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.615-618
• /
• 1999

In this paper, a object detection and tracking algorithm is presented which exhibits robust properties for image sequences with complex background. The proposed algorithm is composed of three parts: moving object detection, object tracking, and motion analysis. The moving object detection algorithm is implemented using a temporal median background method which is suitable for real-time applications. In the motion analysis, we propose a new technique for removing a temporal clutter, such as a swaying plant or a light reflection of a background object. In addition, we design a multiple vehicle tracking system based on Kalman filtering. Computer simulation of the proposed scheme shows its robustness for MPEG-7 test image sequences.

• Journal of Broadcast Engineering
• /
• v.13 no.5
• /
• pp.681-684
• /
• 2008

We proposed a method to improve moving object detection capability of Gaussian Mixture Model by suggesting shape adaptive bidirectional block matching algorithm. This method achieves more accurate detection and tracking performance at various motion types such as slow, fast, and bimodal motions than that of Gaussian Mixture Model. Experimental results showed that the proposed method outperformed the conventional methods.

• Journal of the Korean Society for Precision Engineering
• /
• v.24 no.6
• /
• pp.66-71
• /
• 2007

Development of a locomotive mechanism for the capsule type endoscopes will largely enhance their ability to diagnose disease of digestive organs. As a part of it, there should be provided a detection device of their position in human organs for the purpose of observation and motion control. In this paper, a permanent magnet outside human body was employed to project magnetic field on a capsule type endoscope, while its position dependent flux density was measured by three hall-effect sensors which were orthogonally installed inside the capsule. In order to detect the 2-D position data of the capsule with three hall-effect sensors including the roll, pitch and yaw angle, the permanent magnet was extra translated during the measurement. In this way, the 2-D coordinates and three rotation angles of a capsule endoscope on the same motion plane with the permanent magnet could be detected. The working principle and performance test results of the capsule position detection device were introduced in this paper showing that they could be also applied to 6-DOF position detection.

• Journal of Korea Multimedia Society
• /
• v.22 no.4
• /
• pp.432-442
• /
• 2019

Most current human action recognition methods based on deep learning methods. It is required, however, a very high computational cost. In this paper, we propose an action change detection method to reduce repetitive human action recognition tasks. In reality, simple actions are often repeated and it is time consuming process to apply high cost action recognition methods on repeated actions. The proposed method decides whether action has changed. The action recognition is executed only when it has detected action change. The action change detection process is as follows. First, extract the number of non-zero pixel from motion history image and generate one-dimensional time-series data. Second, detecting action change by comparison of difference between current time trend and local extremum of time-series data and threshold. Experiments on the proposed method achieved 89% balanced accuracy on action change data and 61% reduced action recognition repetition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.6
• /
• pp.2444-2463
• /
• 2020

Significant detection of video can more rationally allocate computing resources and reduce the amount of computation to improve accuracy. Deep learning can extract the edge features of the image, providing technical support for video saliency. This paper proposes a new detection method. We combine the Convolutional Neural Network (CNN) and the Deep Bidirectional LSTM Network (DB-LSTM) to learn the spatio-temporal features by exploring the object motion information and object motion information to generate video. A continuous frame of significant images. We also analyzed the sample database and found that human attention and significant conversion are time-dependent, so we also considered the significance detection of video cross-frame. Finally, experiments show that our method is superior to other advanced methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.4
• /
• pp.807-821
• /
• 1996

In this paper, we proposed a nonlinear hierarchical mtion estimation method. Generally, the conventional hierarchical motion estimation methods have been proposed for fast convergence and detection of large motions. But they have a common drawback that large error in motion estimation is propapated across motion discontinuities. This artifiact is due to the constriaint of motion continuity and the linear interpolation of motion vectors between hierarchical levels. In this paper, we propose an effective hierarchical motion estimation mechod that is robust to motion discontinuities. The proposed algorithm is based on the decomposition of the functional domain for optimizing the intra-level motion estimation functional. Also, we propose an inter-level nonlinear motion estimation equation rather than using the conventional linearprojection scheme of motion field. computer simulations with several test sequences show tht the proposed algorithm performs better than several conventional methods.

• Journal of Korea Multimedia Society
• /
• v.8 no.3
• /
• pp.345-354
• /
• 2005

Since motion estimation and motion compensation methods remove the redundant data to employ the temporal redundancy in images, it plays an important role in digital video compression. Because of its high computational complexity, however, it is difficult to apply to high-resolution applications in real time environments. If we have a priori knowledge about the motion of an image block before the motion estimation, the location of a better starting point for the search of an exact motion vector can be determined to expedite the searching process. In this paper presents the motion detection algorithm that can run robustly about recusive motion. The motion detection compares and analyzes two frames each other, motion of whether happened judge. Through experiments, we show significant improvements in the reduction of the computational time in terms of the number of search steps without much quality degradation in the predicted image.