• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1434-1449
• /
• 2019

Video object segmentation is a significant task in computer vision, but its performance is not very satisfactory. A method of video object segmentation using weakly temporal information is presented in this paper. Motivated by the phenomenon in reality that the motion of the object is a continuous and smooth process and the appearance of the object does not change much between adjacent frames in the video sequences, we use a feed-forward architecture with motion estimation to predict the mask of the current frame. We extend an additional mask channel for the previous frame segmentation result. The mask of the previous frame is treated as the input of the expanded channel after processing, and then we extract the temporal feature of the object and fuse it with other feature maps to generate the final mask. In addition, we introduce multi-mask guidance to improve the stability of the model. Moreover, we enhance segmentation performance by further training with the masks already obtained. Experiments show that our method achieves competitive results on DAVIS-2016 on single object segmentation compared to some state-of-the-art algorithms.

• International Journal of Computer Science & Network Security
• /
• 제24권3호
• /
• pp.151-159
• /
• 2024

Twin clutch model enables the power-shifts as conventional planetary automatic transmission and eradicates the disadvantages of single clutch trans- mission. The automatic control of the dual clutches is a problem. Particularly to control the clutching component that engages when running in one direction of revolution and disengages when running the other direction, which exchange the torque smoothly during torque phase of the gearshifts on planetary-type automatic transmissions, seemed for quite a while hard to compensate through clutch control. Another problem is to skip gears during multiple gearshifts. However, the twin clutch gear control described in ["M Goetz, M C Levesley and D A Crolla. Dynamics and control of gearshifts on twin clutch transmissions, Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 2005"], a significant improvement in twin clutch gear control system is discussed. In this research our objective is to formally specify the twin clutch gear control system and verify it with the help of formal methods. Formal methods have a high potential to give correctness estimating techniques. We use UPPAAL for formal specification and verification. Our results show that the twin clutch gear control model partially fulfills its functional requirements.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권3호
• /
• pp.1105-1118
• /
• 2014

In this paper, a new reversible data hiding method based on a dual binary tree of embedding levels is proposed. Four neighborhood pixels in the upper, below, left and right of each pixel are used as reference pixels to estimate local complexity for deciding embeddable and non-embeddable pixels. The proposed method does not need to record pixels that might cause underflow, overflow or unsuitable for embedment. This can reduce the size of location map and release more space for payload. Experimental results show that the proposed method is more effective in increasing payload and improving image quality than some recently proposed methods.

• Journal of electromagnetic engineering and science
• /
• 제9권4호
• /
• pp.188-193
• /
• 2009

A K-band low-power miniaturized planar-type hyperthermia system was developed to replace massive and expensive equipment. The system consists of a VCO with a buffer amplifier, a high-power amplifier module, a 20-dB-coupled line coupler, a chip circulator and two power detectors for signal generation, amplification and power monitoring. All these components have been implemented in planar form on two module blocks. The total size of the hyperthermia system was less than $10\times6.5\times3\;cm^3$. In order to verify the system performance, ablations were carried out on nude mice xenografted with human breast cancer. Ablation results show performance comparable to the massive components-based system. This work shows the feasibility of a low-cost miniaturized hyperthermia system for practical clinical applications.

• 통합자연과학논문집
• /
• 제11권3호
• /
• pp.148-153
• /
• 2018

In this paper, we propose a hybrid-recognition system to obtain the state information and control the Internet of Things (IoT) based smart devices using two recognitions. First, we use a facial recognition for checking the owner of the mobile devices, i.e., smartphones, tablet PCs, and so on, and obtaining the state information of the IoT based smart devices, i.e., smart cars, smart appliance, and so on, and then we use a fingerprint recognition to control them. Further, in the conventional system, the message of the state and control information between the mobile devices and smart devices is only exchanged through the cellar mobile network. Thus, we also propose a direct communication to reduce the total transmission time. In addition, we develop a testbed of the proposed system using smartphones, desktop computers, and Arduino vehicle as one of the smart devices. We evaluate the total transmission time between the conventional and direct communications and show that the direct communication with the proposed system has better performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권10호
• /
• pp.2650-2662
• /
• 2012

Due to the semantic gap issue, the performance of automatic image annotation is still far from satisfactory. Active learning approaches provide a possible solution to cope with this problem by selecting most effective samples to ask users to label for training. One of the key research points in active learning is how to select the most effective samples. In this paper, we propose a novel active learning approach based on sparse graph. Comparing with the existing active learning approaches, the proposed method selects the samples based on two criteria: uncertainty and representativeness. The representativeness indicates the contribution of a sample's label propagating to the other samples, while the existing approaches did not take the representativeness into consideration. Extensive experiments show that bringing the representativeness criterion into the sample selection process can significantly improve the active learning effectiveness.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1113-1135
• /
• 2018

Existing trust evaluation models for wireless sensor networks can accurately and objectively evaluate trust value of nodes, but the nodes' energy saving problem was ignored. Especially when there are a few malicious nodes in a network, the overall trust value calculation for all nodes would waste lots of energy. Beside that, the network failure caused by nodes death was also not considered. In this paper, we proposed a method for avoiding energy hole which applied trust evaluation models and a trust evaluation method based on information entropy, so as to achieve the purpose of improving nodes utilization. Simulation results show that the proposed method can effectively improve nodes utilization, and it has reasonable detection rate and lower false alert rate compared to other classical methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.490-490
• /
• 2000

Parallel genetic algorithms are particularly easy to implement and promise substantial gains in performance. Its basic idea is to keep several subpopulations that are processed by genetic algorithms. Furthermore, a migration mechanism produces a chromosome exchange between subpopulation. In this paper, a new selection method based on non-linear fitness assignment presented. The use of proposed ranking selection permits higher local exploitation search, where the diversity of populations is structure. Experimental results show that the relation between local-global search balance and the probabilities of reaching a desired solution.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권6호
• /
• pp.1113-1132
• /
• 2011

In wireless sensor networks (WSNs), most energy saving asynchronous MAC protocols are custom tailored for unicast communications only. However, broadcast protocols are very commonly used in WSNs for a variety of functionalities, such as gathering network topology information, event monitoring and query processing. In this paper, we propose a novel low-power asynchronous broadcast MAC protocol called Alarm Broadcast (A-CAST). A-CAST employs the strobe preamble that specifies the residual waiting time for the following data transmission. Each receiver goes back to sleep upon hearing the strobe preamble for the residual time duration, to conserve energy and to wake up just before data transmission starts. We compute the energy consumption of A-CAST via rigorous mathematical analysis. The analytic results show that A-CAST outperforms B-CAST, a simple broadcast extension of the well-known B-MAC. We also implement A-CAST on sensor motes and evaluated its performance through real experiments. Our experimental results show that A-CAST reduces the energy consumption by up to 222% compared to the previously proposed protocols.

• 대한의용생체공학회:의공학회지
• /
• 제30권3호
• /
• pp.199-204
• /
• 2009

Iridium oxide films (IROFs) have been extensively studied as a material for electrical stimulation of neurons, as iridium oxide has higher charge storage capacity than other metal films. More recently, sputtered iridium oxide film (SIROF) has been studied, because it can be made more conveniently than activated iridium oxide film (AIROF). Typically, the SIROFs are grown at temperatures from 400 to 600 $^{\circ}C$. However, such high temperatures cannot be used when the iridium oxide (IrOx) film is to be deposited on a flexible polymer material, such as polyimide. In this paper, we show that we can still obtain excellent characteristics in SIROFs grown without heating (cold SIROF), by optimizing the growth conditions. We show that the oxygen flow rate is a critical parameter for controlling the surface properties of a cold SIROF. At an oxygen flow rate of 12 seem, the cold SIROF exhibited a charge storage capacity (CSC) of 60 mC/cm$^2$, which is comparable to or better than other published values for iridium oxide films including heated SIROFs. The film produced under these conditions also had the minimum impedance value of all cold SIROFs deposited for this study. A stability test and biocompatibility test also demonstrated the superiority of the optimized cold SIROF.