• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.6
• /
• pp.313-321
• /
• 2021

The global market for medical devices is huge, and it will continue showing remarkable growth in the future. However, due to the entry barrier to develop medical devices, many domestic companies have technical problems in entering the medical device industry. In this paper, we introduce an open platform that can help with research and development for companies in the healthcare industry. This open platform consists of a hardware part and a software part. A hardware part is combined into CPU, base and other modules that are easy to replace and assemble. A software part is based on application software for development developed by Bionet. We test the performance of the open medical device platform using a biosignal processing algorithm.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• Journal of information and communication convergence engineering
• /
• v.19 no.3
• /
• pp.188-197
• /
• 2021

In military operations, an accurate localization system is required to navigate soldiers to their destinations, even in non-GPS environments. The global positioning system is a commonly used localization method, but it is difficult to maintain the robustness of GPS-based localization against jamming of signals. In addition, GPS-based localization cannot provide important terrain information such as obstacles. With the widespread use of embedded sensors, sensor-based pedestrian tracking schemes have become an attractive option. However, because of noisy sensor readings, pedestrian tracking systems using motion sensors have a major drawback in that errors in the estimated displacement accumulate over time. We present a group-based standalone system that creates terrain maps automatically while also locating soldiers in mountainous terrain. The system estimates landmarks using inertial sensors and utilizes split group information to improve the robustness of map construction. The evaluation shows that our system successfully corrected and combined the drift error of the system localization without infrastructure.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.9
• /
• pp.3384-3400
• /
• 2021

With the advent of the data era, people pay much more attention to data corruption. Aiming at the problem that the majority of existing schemes do not support corruption detection of ciphertext data stored in cloud environment, this paper proposes a data corruption detection scheme based on ciphertexts in cloud environment (DCDC). The scheme is based on the anomaly detection method of Gaussian model. Combined with related statistics knowledge and cryptography knowledge, the encrypted detection index for data corruption and corruption detection threshold for each type of data are constructed in the scheme according to the data labels; moreover, the detection token for data corruption is generated for the data to be detected according to the data labels, and the corruption detection of ciphertext data in cloud storage is realized through corresponding tokens. Security analysis shows that the algorithms in the scheme are semantically secure. Efficiency analysis and simulation results reveal that the scheme shows low computational cost and good application prospect.

• The Journal of Korea Robotics Society
• /
• v.16 no.3
• /
• pp.270-275
• /
• 2021

Most of the initial forms of cooperative robots were intended to repeat simple tasks in a given space. So, they showed no significant difference from industrial robots. However, research for improving worker's productivity and supplementing human's limited working hours is expanding. Also, there have been active attempts to use it as a service robot by applying AI technology. In line with these social changes, we produced a mobile manipulator that can improve the worker's efficiency and completely replace one person. First, we combined cooperative robot with mobile robot. Second, we applied speech recognition technology and deep learning based object detection. Finally, we integrated all the systems by ROS (robot operating system). This system can communicate with workers by voice and drive autonomously and perform the Pick & Place task.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1431-1437
• /
• 2018

Securing objects in the Internet of Things (IoT) is essential. Authentication model is one candidate to secure an object, but it is only limited to handle a specific type of attack such as Sybil attack. The authentication model cannot handle other types of attack such as trust-based attacks. This paper proposed two-phase security protection for objects in IoT. The proposed method combined authentication and statistical models. The results showed that the proposed method could handle other attacks in addition to Sybil attacks, such as bad-mouthing attack, good-mouthing attack, and ballot stuffing attack.

• Journal of Information Processing Systems
• /
• v.15 no.2
• /
• pp.344-358
• /
• 2019

Surveillance cameras have installed in many places because security and safety is becoming important in modern society. Through surveillance cameras installed, we can deal with troubles and prevent accidents. However, watching surveillance videos and judging the accidental situations is very labor-intensive. So now, the need for research to analyze surveillance videos is growing. This study proposes an algorithm to track multiple persons using SURF and background subtraction. While the SURF algorithm, as a person-tracking algorithm, is robust to scaling, rotating and different viewpoints, SURF makes tracking errors with sudden changes in videos. To resolve such tracking errors, we combined SURF with a background subtraction algorithm and showed that the proposed approach increased the tracking accuracy. In addition, the background subtraction algorithm can detect persons in videos, and SURF can initialize tracking targets with these detected persons, and thus the proposed algorithm can automatically detect the enter/exit of persons.

• Earthquakes and Structures
• /
• v.16 no.4
• /
• pp.425-436
• /
• 2019

Successive ground motions having short time intervals have occurred in many earthquakes so far. It is necessary to investigate the effects of this phenomenon on different types of structures and to take these effects into consideration while designing or retrofitting structures. The effects of seismic sequences on the structures with combined reinforced concrete shear wall and moment resisting frame system have not been investigated in details yet. This paper has tried to analyse the seismic performance of structures with such structural systems subjected to mainshock-aftershock sequences. The effects of the seismic sequences on the investigated models are evaluated by strong measures such as IDA capacity and fragility and vulnerability curves. The results of this study show that the seismic sequences have a significant effect on the investigated models, which necessitates considering this effect on designing, retrofitting, decision making, and taking precautions.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.12
• /
• pp.145-151
• /
• 2018

While commercialization of IoT technologies in the safety management sector is being promoted in terms of industrial safety of large indoor businesses, implementing a system for risk management of small outdoor work sites with frequent site movements is not actively implemented. In this paper, we propose an efficient dynamic workload balancing strategy which combined low-power, wide-bandwidth (LPWA) communication and low-power Bluetooth (BLE) communication technologies to support customized risk management alarm systems for each individual (driver/operator/manager). This study was designed to enable long-term low-power collection and transmission of traffic information in outdoor environment, as well as to implement an integrated real-time safety management system that notifies a whole field worker who does not carry a separate smart device in advance. Performance assessments of the system, including risk alerts to drivers and workers via Bluetooth communication, the speed at which critical text messages are received, and the operation of warning/lighting lamps are all well suited to field application.