• The Journal of the Korea institute of electronic communication sciences
• /
• v.12 no.2
• /
• pp.331-336
• /
• 2017

The need for early fire detection technology is increasing in order to prevent fire disasters. Sensor device detection for heat, smoke and fire is widely used to detect flame and smoke, but this system is limited by the factors of the sensor environment. To solve these problems, many image-based fire detection systems are being developed. In this paper, we implemented a system to detect fire and smoke from camera input images using a convolution neural network. Through the implemented system using the convolution neural network, a feature map is generated for the smoke image and the fire image, and learning for classifying the smoke and fire is performed on the generated feature map. Experimental results on various images show excellent effects for classifying smoke and fire.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.10 no.1
• /
• pp.55-60
• /
• 2010

For battery-powered device applications, which grow rapidly in the electronic market today, low-power becomes one of the most important design issues of CMOS VLSI circuits. A multi-VDD system, which uses more than one power-supply voltage in the same system, is an effective way to reduce the power consumption without degrading operating speed. However, in the multi-VDD system, level converters should be inserted to prevent a large static current flow for the low-to-high conversion. The insertion of the level converters induces the overheads of power consumption, delay, and area. In this paper, we propose a new level converter which can provide the level up/down conversions for the various input and output voltages. Since the proposed level converter uses only one power-supply voltage, it has an advantage of reducing the complexity in physical design. In addition, the proposed level converter provides lower power and higher speed, compared to existing level converters.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.15 no.3
• /
• pp.365-373
• /
• 2015

We propose a simple and compact thyristor-based neuron circuit. The thyristor exhibits bi-stable characteristics that can mimic the action potential of the biological neuron, when it is switched between its OFF-state and ON-state with the help of assist circuit. In addition, a method of inter-neuron connection with synaptic devices is proposed, using double current mirror circuit. The circuit utilizes both short-term and long-term plasticity of the synaptic devices by flowing current through them and transferring it to the post-synaptic neuron. The double current mirror circuit is capable of shielding the pre-synaptic neuron from the post synaptic-neuron while transferring the signal through it, maintaining the synaptic conductance unaffected by the change in the input voltage of the post-synaptic neuron.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.4
• /
• pp.291-296
• /
• 2015

This paper describes the design of a high-speed comparator for high-speed automatic test equipment (ATE). The normal comparator block, which compares the detected signal from the device under test (DUT) to the reference signal from an internal digital-to-analog converter (DAC), is composed of a rail-to-rail first pre-amplifier, a hysteresis amplifier, and a third pre-amplifier and latch for high-speed operation. The proposed continuous comparator handles high-frequency signals up to 800MHz and a wide range of input signals (0~5V). Also, to compare the differences of both common signals and differential signals between two DUTs, the proposed differential mode comparator exploits one differential difference amplifier (DDA) as a pre-amplifier in the comparator, while a conventional differential comparator uses three op-amps as a pre-amplifier. The chip was implemented with $0.18{\mu}m$ Bipolar CMOS DEMOS (BCDMOS) technology, can compare signal differences of 5mV, and operates in a frequency range up to 800MHz. The chip area is $0.514mm^2$.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.11
• /
• pp.58-66
• /
• 1998

A low power consuming laser scanning vibrometer is studied for its development. For its optical system, a laser interferometer is constructed to use the Doppler effect. In order to reduce the driving power of the scanning system, a small displacement of the scanning system is produced, which is achieved by using a magnetostrictive actuator. A sufficient rotating angle of the scanning system is obtained by using an amplified displacement from the resonant phenomena of a second order mechanical system composed of a mass and spring. The control of the magnetostrictive actuator using a Terfenol-D is performed without using a feedback system to help reduce the power consumption. The vibration analysis is made for the sinusoidal scanning input to have the space domain information from the time domain of the velocity of a vibration object. As a partial work of development of a tow power consuming laser scanning vibrometer, in this work, a scanning system which has the above features is developed and experimentally investigated. For the purpose of the optical system calibration, the vibration measurement for one axis is presented and the future works are discussed.

• Proceedings of the KIPE Conference
• /
• 2018.11a
• /
• pp.62-64
• /
• 2018

In two-stage single-phase inverters, inherent double line frequency component is present at both input and output of the front-end converter. Generally large electrolytic capacitors are required to eliminate the ripple. It is well known that the low frequency ripple shortens the lifespan of the capacitor hence the system reliability. However, the ripple can hardly be eliminated without the hardware combined with an energy storage device or a certain control algorithm. In this paper, a novel power-decoupling control method is proposed to eliminate the double line frequency ripple at the front-end converter of the DC/AC power conversion system. The proposed control algorithm is composed of two loop, ripple rejection loop and average voltage control loop and no extra hardware is required. In addition, it does not require any information from the phase-locked-loop (PLL) of the inverter and hence it is independent of the inverter control. In order to prove the validity and feasibility of the proposed algorithm a 5kW Dual Active Bridge DC/DC converter and a single-phase inverter are implemented, and experimental results are presented.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.3
• /
• pp.85-90
• /
• 2021

In this paper, fall detection using individual wearable devices for older people is considered. To design a low-cost wearable device for reliable fall detection, we present a comprehensive analysis of two representative models. One is a machine learning model composed of a decision tree, random forest, and Support Vector Machine(SVM). The other is a deep learning model relying on a one-dimensional(1D) Convolutional Neural Network(CNN). By considering data segmentation, preprocessing, and feature extraction methods applied to the input data, we also evaluate the considered models' validity. Simulation results verify the efficacy of the deep learning model showing improved overall performance.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.4
• /
• pp.23-28
• /
• 2021

Augmented reality is a technique for combining virtual images into real life by showing information of virtual 3D objects on top of a real-world environment (Azuma et al., 2001). This study is an augmented reality-based educational content delivery device that receives user input that selects either a preset object or a photographed object for augmented reality-based training; It includes a three-dimensional design generation unit that generates a stereoscopic model of the augmented reality environment from an object, a three-dimensional view of the scene, a disassembly process of the developing road from a three-dimensional model, and a content control unit provided by the user terminal by generating educational content including a three-dimensional model, a scene chart, a scene, a decomposition process, and a coupling process to build a coupling process from the scene to the three-dimensional model in an augmented reality environment. The next study provides a variety of educational content so that children can use AR technology as well as shapes to improve learning effectiveness. We also believe that studies are needed to quantitatively measure the efficacy of which educational content is more effective when utilizing AR technology.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.4
• /
• pp.191-197
• /
• 2022

This study proposes an image-based Pose Intention Network (PIN) algorithm for rehabilitation via patients' intentions. The purpose of the PIN algorithm is for enabling an active rehabilitation exercise, which is implemented by estimating the patient's motion and classifying the intention. Existing rehabilitation involves the inconvenience of attaching a sensor directly to the patient's skin. In addition, the rehabilitation device moves the patient, which is a passive rehabilitation method. Our algorithm consists of two steps. First, we estimate the user's joint position through the OpenPose algorithm, which is efficient in estimating 2D human pose in an image. Second, an intention classifier is constructed for classifying the motions into three categories, and a sequence of images including joint information is used as input. The intention network also learns correlations between joints and changes in joints over a short period of time, which can be easily used to determine the intention of the motion. To implement the proposed algorithm and conduct real-world experiments, we collected our own dataset, which is composed of videos of three classes. The network is trained using short segment clips of the video. Experimental results demonstrate that the proposed algorithm is effective for classifying intentions based on a short video clip.

• Journal of the Korean Society for Precision Engineering
• /
• v.13 no.8
• /
• pp.112-119
• /
• 1996

Among the various type of accelerometer, the servo rebalancing type accelermoter can be suitable for Inertial Navigation System, because of its high sensitivity and good response in low frequency. In this paper, we proposed a new technology to control inductive tuype accelerometer utilizing digital PWM method. The new developed digital PWM control has special design scheme for transmitting measurement value to outer device in its servo ollp. So it has no quantized error of transforming outputs of sensors to digital domain. The quantized error may make serious problem in INS system, because outputs of sensor are integrated once or twice by digital computer and it happens every sensor reading times. Therefore, in order to get the accurate information such as displacement, it is necessary to measure accurately the input current. In addition, Digital Signal Processing needs digital data transmission, digital PWM method is adaptive for this purpose. We realized a practical circuit for digital PWM control, analyzed the stability of the circuit, and designed the controller etc. In this study, we solved many practical problem for this application, and got out good results.