• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.11
• /
• pp.5491-5505
• /
• 2017

Use of the Gaussian inverse Wishart PHD (GIW-PHD) filter has demonstrated promise as an approach to track an unknown number of extended targets. However, the partitioning approaches used in the GIW-PHD filter, such as distance partition with sub-partition (DP-SP), prediction partition (PP) and expectation maximization partition (EMP), fails to provided accurate partition results when targets are spaced closely together and performing maneuvers. In order to improve the performance of a GIW-PHD filter, this paper presents a cooperation partitioning (CP) algorithm to solve the partitioning issue when targets are spaced closely together. In the GIW-PHD filter, the DP-SP is insensitive to target maneuvers but sensitive to the differences in target sizes, while EMP is the opposite. The proposed CP algorithm is a fusion approach of DP-SP and EMP, which employs EMP as a sub-partition approach after DP. Therefore, the CP algorithm will be sensitive to neither target maneuvers nor differences in target sizes. The simulation results show that the use of the proposed CP algorithm will improve the performance of the GIW-PHD filter when targets are spaced closely together.

• Journal of Biomedical Engineering Research
• /
• v.9 no.1
• /
• pp.101-108
• /
• 1988

In this paper the power spectrum of background EEG is estimated by the LMS PHD based on least mean square. At the power spectrum estimatiom, the stocastic process of background EEG is assumed to consist of the nonharmonic sinusoid and the white noise. In the LMS PHD the model parameters are obtained by the least mean square at optimal order which is obtained from the fact that the eigenvalue's fluctuation of autocorrelation matrix of the normal back-ground EEG is smaller at some order than at other order when the power spectrum of background EEG is esitmated by PHD. The optimal order of this model is the 6-th order when the eigenvalue's fluctuation of autocorrelation matrix of background EEG is considered. The estimation results are with compared the results from the Maximum Entropy Spectral Estimation and Pisarenko Harmonic Decomposition. From the comparison results. The LMS PHD is possible to estimate the power spectrum of background EEG.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.12
• /
• pp.150-157
• /
• 2015

Pedestrian detection has largely been researched as one of the important technologies for autonomous driving vehicle and preventing accidents. There are two categories for pedestrian detection, camera-based and LIDAR-based. LIDAR-based methods have the advantage of the wide angle of view and insensitivity of illuminance change while camera-based methods have not. However, there are several problems with 3D LIDAR, such as insufficient resolution to detect distant pedestrians and decrease in detection rate in a complex situation due to segmentation error and occlusion. In this paper, two methods using GM-PHD filter are proposed to improve the poor rates of pedestrian detection algorithms based on 3D LIDAR. First one improves detection performance and resolution of object by automatic accumulation of points in previous frames onto current objects. Second one additionally enhances the detection results by applying the GM-PHD filter which is modified in order to handle the poor situation to classified multi target. A quantitative evaluation with autonomously acquired road environment data shows the proposed methods highly increase the performance of existing pedestrian detection algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.4C
• /
• pp.307-312
• /
• 2012

In this paper, we perform a complexity analysis for implementation of the ISO/IEEE 11073 Personal Health Device (PHD) standards in order to check the required system resources when ISO/IEEE 11073 PHD standards are implemented on the embedded system. Base on the implemented programs complying the PHD standards for a weighing scale, a blood pressure monitor, and a glucose meter among the various personal health devices, we make a pseudo-code. And then from the two different points of view such as program memory space and data memory space, we make a complexity analysis model. Because system resources or capability are strongly restricted in the personal health devices, our research work is very useful to estimate the required system resources.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.9
• /
• pp.773-777
• /
• 2016

In this paper, a novel multi-object tracking method to track an unknown number of objects is proposed. To handle multiple object states and uncertain observations efficiently, a probability hypothesis density (PHD) filter is adopted and modified. The PHD filter is capable of reducing false positives, managing object appearances and disappearances, and estimating the multiple object trajectories in a unified framework. Although the PHD filter is robust in cluttered environments, it is vulnerable to false negatives. For this reason, we propose to exploit local observations in an RFS of the observation model. Each local observation is generated by using an online trained object detector. The main purpose of the local observation is to deal with false negatives in the PHD filtering procedure. The experimental results demonstrated that the proposed method robustly tracked multiple objects under practical situations.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.9
• /
• pp.1657-1665
• /
• 2016

Various fields are being researched through Deep Learning using CNN(Convolutional Neural Network) and these researches show excellent performance in the image recognition. In this paper, we provide streamlined GoogLeNet of CNN architecture that is capable of learning a large-scale Korean character database. The experimental data used in this paper is PHD08 that is the large-scale of Korean character database. PHD08 has 2,187 samples for each character and there are 2,350 Korean characters that make total 5,139,450 sample data. As a training result, streamlined GoogLeNet showed over 99% of test accuracy at PHD08. Also, we made additional Korean character data that have fonts that are not in the PHD08 in order to ensure objectivity and we compared the performance of classification between streamlined GoogLeNet and other OCR programs. While other OCR programs showed a classification success rate of 66.95% to 83.16%, streamlined GoogLeNet showed 89.14% of the classification success rate that is higher than other OCR program's rate.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.5
• /
• pp.464-470
• /
• 2013

These days, with increasing interest about the health care, research about standard for transmitting/receiving data of health device has been actively conducted. Standardization and testing / certification is underway through the IEEE 11073 PHD, Health Level Seven and the Continua Health Alliance, it'll continue to be more research and experiments in the future. In this paper, we introduce the IEEE 11073 PHD standards for exchange of remote health information and the Bluetooth Health Device Profile(HDP) for health device profile of the Bluetooth technology, also we conducted experiments about exchange health information between health device and smart device which are adapted the IEEE 11073 PHD and Bluetooth HDP. Transmitting/receiving of the IEEE 11073 PHD standard protocol using the Bluetooth HDP profile between health device and smart device can be known by this experiment, and it expect to implement variety of remote medical services through health device supporting these standards in future.

• ETRI Journal
• /
• v.38 no.5
• /
• pp.1019-1029
• /
• 2016

The probability hypothesis density (PHD) filter is an effective means to track multiple targets in that it avoids explicit data associations between the measurements and targets. However, the target birth intensity as a prior is assumed to be known before tracking in a traditional target-tracking algorithm; otherwise, the performance of a conventional PHD filter will decline sharply. Aiming at this problem, a novel target birth intensity scheme and an improved measurement-driven scheme are incorporated into the PHD filter. The target birth intensity estimation scheme, composed of both PHD pre-filter technology and a target velocity extent method, is introduced to recursively estimate the target birth intensity by using the latest measurements at each time step. Second, based on the improved measurement-driven scheme, the measurement set at each time step is divided into the survival target measurement set, birth target measurement set, and clutter set, and meanwhile, the survival and birth target measurement sets are used to update the survival and birth targets, respectively. Lastly, a Gaussian mixture implementation of the PHD filter is presented under a linear Gaussian model assumption. The results of numerical experiments demonstrate that the proposed approach can achieve a better performance in tracking systems with an unknown newborn target intensity.

• Journal of IKEEE
• /
• v.19 no.3
• /
• pp.447-454
• /
• 2015

In this paper, the complexity analysis for implementation of the PM-store is performed in terms of the number of instruction cycles which is executed by CPU in a personal health device(PHD) in order to transfer the large amount of the periodically generated measurement data using the PM-store concept defined in ISO/IEEE 11073 PHD standards. We propose an analytic model that is focused on the number of instruction cycles executed by CPU depending on the PM-store hierarchy.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.97-100
• /
• 2016

Using the CNN(Convolutinal Neural Network), Deep Learning for variety of fields are being developed and these are showing significantly high level of performance at image recognition field. In this paper, we show the test accuracy which is learned by large-scale training data, over 5,000,000 of Korean characters. The architecture of CNN used in this paper is KCR(Korean Character Recognition)-AlexNet newly created based on AlexNet. KCR-AlexNet finally showed over 98% of test accuracy. The experimental data used in this paper is large-scale Korean character database PHD08 which has 2,187 samples for each Korean character and there are 2,350 Korean characters that makes total 5,139,450 sample data. Through this study, we show the excellence of architecture of KCR-AlexNet for learning PHD08.