• Journal of KIISE
• /
• v.41 no.12
• /
• pp.1145-1151
• /
• 2014

In this paper, we propose MPLO(Multi-Path routing with Local Optimization) for load balancing in the tactical backbone network. The MPLO manages global metric and local metric separately. The global metric is propagated to other routers via a routing protocol and is used for configuring loop-free multi-path. The local metric reflecting link utilization is used to find an alternate path when congestion occurs. We verified MPLO could effectively distribute user traffic among available routers by simulation.

• The KIPS Transactions:PartC
• /
• v.12C no.4 s.100
• /
• pp.543-550
• /
• 2005

Many routing algorithms, proposed for ad-hoc wireless networks, we based on source routing scheme and shortest path route has short lifetime especially in highly dense ad-hoc wireless networks. So some routing protocols such as SSA and ABR are considering the link stability and try finding more stable route. In this paper we propose a new routing algorithm considering utilization metric based on SSA routing algerian in Ad-Hoc networks. To reduce the bottleneck by specific metric of SSA, proposed scheme makes load balancing in networks by distributing the connections to several routes. For the evaluation of the performance we compare our scheme with existent routing protocol AODV and SSA. And the results, obtained using the ns-2 network simulation platform, show good performance that reduced the number of reconstructions remarkably by distributing the whole traffic to several routes when there are several stable routes.

• Journal of Broadcast Engineering
• /
• v.20 no.2
• /
• pp.204-214
• /
• 2015

Depth map upsampling is an approach to increase the spatial resolution of depth maps obtained from a depth camera. Depth map quality is closely related to 3D perception of stereoscopic image, multi-view image and holography. In general, the performance of upsampled depth map is evaluated by PSNR (Peak Signal to Noise Ratio). On the other hand, time-consuming 3D subjective tests requiring human subjects are carried out for examining the 3D perception as well as visual fatigue for 3D contents. Therefore, if an objective metric is closely correlated with a subjective test, the latter can be replaced by the objective metric. For this, this paper proposes a best metric by investigating the relationship between diverse objective metrics and 3D subjective tests. Diverse reference and no-reference metrics are adopted to evaluate the performance of upsampled depth maps. The subjective test is performed based on DSCQS test. From the utilization and analysis of three kinds of correlations, we validated that SSIM and Edge-PSNR can replace the subjective test.

• Journal of KIISE
• /
• v.42 no.8
• /
• pp.1057-1064
• /
• 2015

The tactical network is system based on wireless networking technologies that ties together surveillance reconnaissance systems, precision strike systems and command and control systems. Several alternative paths exist in the network because it is connected as a grid to improve its survivability. In addition, the network topology changes frequently as forces and combatants change their network access points while conducting operations. However, most Internet routing standards have been designed for use in stable backbone networks. Therefore, tactical networks may exhibit a deterioration in performance when these standards are implemented. In this paper, we propose Priority based Multi-Path routing with Local Optimization(PMPLO) for a tactical backbone network. The PMPLO separately manages the global and local metrics. The global metric propagates to other routers through the use of a routing protocol, and it is used for a multi-path configuration that is guaranteed to be loop free. The local metric reflects the link utilization that is used to find an alternate path when congestion occurs, and it is managed internally only within each router. It also produces traffic that has a high priority privilege when choosing the optimal path. Finally, we conducted a simulation to verify that the PMPLO can effectively distribute the user traffic among available routers.

• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.353-356
• /
• 2019

The main functionality of occupancy sensors is to determine the existence of humans in the space. If the space is occupied, a light is on and for vacancy, the light automatically turns off. In this letter, the functionality is realized by the utilization of color information. The color information of incoming people is saved. For outgoing people, their color distribution is compared with the saved information, thus providing the recognition of the outgoing people. For the comparison, four similarity metrics are examined to validate the proposed method.

• Journal of Broadcast Engineering
• /
• v.17 no.3
• /
• pp.459-470
• /
• 2012

In this paper, we propose a method to upsample a low-resolution depth map to a high-resolution version. While conventional camera sensors produce high-resolution color images, the sizes of the depth maps of range/depth sensors are usually low. In this paper, we consider the utilization of high-frequency components to the conventional depth map interpolation methods such as bilinear, bicubic, and bilateral. The proposed method is composed of the three steps: high-frequency component extraction, high-frequency component application, and interpolation. Two objective evaluation measures such as sharpness degree and blur metric are used to examine the performance. Experimental results show that the proposed method significantly outperforms other conventional methods by a factor of 2 in terms of sharpness degree. As well, a blur metric is reduced by a factor of 14 %.

• The KIPS Transactions:PartD
• /
• v.10D no.4
• /
• pp.689-696
• /
• 2003

Criticality prediction models that determine whether a design entity is fault-prone or non fault-prone are used for identifying trouble spots of software system in analysis or design phases. Many criticality prediction models for identifying fault-prone modules using complexity metrics have been suggested. But most of them need training data set. Unfortunately very few organizations have their own training data. To solve this problem, this paper builds a new prediction model, KSM, based on Kohonen SOM neural networks. KSM is implemented and compared with a well-known prediction model, BackPropagation neural network Model (BPM), considering internal characteristics, utilization cost and accuracy of prediction. As a result, this paper shows that KSM has comparative performance with BPM.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39B no.12
• /
• pp.842-855
• /
• 2014

Recently, the progress of smart phone market has retarded by oversupply therefore wearable computer has been the focus of new growth engine. Wearable computing system is a complex fusion of a variety of technologies such as wireless network, embedded, sensor and new material. Because these technologies involves utilization and mobility in addition to quality characteristic in existing software, application of ISO/IEC 9126 is not perfect when assessing quality of wearable computer. In this study, author suggested new quality assessment model for wearable computer by sorting quality attribute in ISO/IEC 9126 and adding new quality attribute. For this, author investigated features and functional requirements related to wearable computer. and then author suggested quality standard and metrics by identifying quality characteristic. Author confirmed practicality of quality assessment model by using suggested model in scenario and comparing quality assessment of three goods such as company S, L, G. This quality assessment model is expected to use guidelines for assessing quality of wearable computer.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.2
• /
• pp.266-274
• /
• 2021

Ship officers are subject to excessive job stress, which has negative physical and psychological impacts and may adversely affect the smooth supply and demand of human resources. In this study, a mobile web application was developed as a tool for systematic job stress measurement and management of officers and verified through quality evaluation. Requirement analysis was performed by ship officers and staff in charge of human resources of shipping companies, and the results were reflected in the application configuration step. The application was designed according to the waterfall model, which is a traditional software development method, and functions were implemented using JSP and Spring Framework. Performance evaluation on the user interface, confirmed that proper input and output results were implemented, and the respondent results and the database were configured in the administrator interface. The results of evaluation questionnaires for quality evaluation of the interface based on ISO/IEC 9126-2 metric were significant 4.60 for the user interface and 4.65 for the administrator interface in a 5-point scale. In the future, it is necessary to conduct follow-up research on the development of data analysis system through utilization of the collected big-data sets.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.