• Journal of the Semiconductor & Display Technology
• /
• v.19 no.2
• /
• pp.88-91
• /
• 2020

Image segmentation is the task of partitioning an image into multiple sets of pixels based on some characteristics. The objective is to simplify the image into a representation that is more meaningful and easier to analyze. In this paper, we apply deep-learning to pre-train the learning model, and implement an algorithm that performs image segmentation in real time by extracting frames for the stream input from the Android device. Based on the open source of DeepLab-v3+ implemented in Tensorflow, some convolution filters are modified to improve real-time operation on the Android platform.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.2044-2045
• /
• 2007

코로나 발생은 전극의 첨점이 형성되어 전극에 전계가 집중될 경우 불균질성에 의한 방전이다. 수용가용 154kV 인류형 클램프 말단의 전선처리는 코로나 발생을 가속시켜 전력손실, 노이즈 발생 등의 영향을 미친다. 본 연구에서는 수용가용 154kV 수전설비용 자기애자 (Porcelain Insulator)와 가공선로(Overhead Line)의 지지용 금구사이에서 발생하는 코로나 특성을 고찰하였다, 일반적 방법으로 확인 되지않는 코로나를 UV Image 이용하여 애자의 소손 위험성 및 상태를 제시하였으며, 유한요소법(FEMLAB)을 이용하여 나전선의 길이별로 전계분포에 의한 위험성을 제시하였다. 수전설비의 설치조건 및 금구류의 유지보수 및 안전성 향상에 활용될 것으로 기대한다.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2003.05c
• /
• pp.120-123
• /
• 2003

To investigate the characteristics of the single dendrimer molecule, we attempt to measure morphology and electrical properties of the self-assembled dendrimer on Au (111)substrate with SPM(scanning probe microscopy). The same self-assembly procedure was used for two different concentrations, $10{\mu}mol/ml$ and $100{\mu}mol/ml$. The case of lower concentration, we can measure the diameter and the height of the single molecule with the tapping mode AFM image. The imaged single molecules were dome shaped and the average diameter and height were 15.6 nm, 1.2 nm respectively. From these sizes, we can calculate the volume of the single molecule. The volume of the single molecule was estimated about $116nm^3$. However, that of higher concentration, it is difficult to obtain obvious image of the single molecule. To add to, I-V characteristics were investigated using STM, on which the phenomenon of negative differential resistance (NDR)was observed between 0.14 V and 0.24 V reproductively.

• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.82-85
• /
• 2004

This paper proposes an advanced circuit for the capacitive type fingerprint sensor signal processing and an effective isolation structure for minimizing an electrostatic discharge(ESD) influence and for removing a signal coupling noise of each sensor pixel. The proposed detection circuit increases the voltage difference between a ridge and valley about $80\%$ more than old circuit. The test chip is composed of $160\;\times\;192$ array sensing cells $(9,913\times11,666\;um^2).$ The sensor plate area is $58\;\times\;58\;um^2$ and the pitch is 60um. The image resolution is 423 dpi. The chip was fabricated on a 0.35um standard CMOS process. It successfully captured a high-quality fingerprint image and performed the registration and identification processing. The sensing and authentication time is 1 sec(.) with the average power consumption of 10 mW at 3.0V. The reveal ESD tolerance is obtained at the value of 4.5 kV.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.6
• /
• pp.935-941
• /
• 2017

This paper proposes a method for estimating the distance betweeen vehicles in a moving situation using the image ratio of the distance between the tail lamps of a front vehicle. The actual distance between the tail lamps of a front vehicle was transmitted by LED tail lamps using visible light communication. As the distance between the front vehicle and the rear vehicle changes, it calculates the ratio of the pixel width between the tail lamps of the front vehicle projected on the image. The calculated values are used to derive a distance-mapping function through non-linear regression technique. Then, the distance between vehicles in the moving situation is estimated based on this function.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.3 s.309
• /
• pp.51-59
• /
• 2006

In this paper, we propose an excellent ultrasonic transceiver system based on a QPSK modulation technique for underwater communication. The transmitter sends a still image at the level of 187dB re $1{\mu}Pa/V@1m$ through a power amplifier by driving an ultrasonic sensor. The receiver performs digital conversion at the 100kHz sampling frequency, demodulation and decoding process for the image sent from the transmitter through the underwater communication. We have shown that the processed image at the receiver is almost the same as the orignal one. The maximum detection distance of the system proposed in this paper is approximately 1.17km. To cope with the difficulties of transmission loss, this paper proposes, implements and analyzes important parameters of sensors and circuits used in the system. Most of the underwater communication has focused on the transmission of audio signal, but this paper suggests an efficient underwater communication system for still image transmission.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.7
• /
• pp.878-883
• /
• 2018

This paper proposes a method for transmitting instruction between vehicles in a moving situation using RC Car having camera. Information of preceding RC Car was transmitted by LED using Optical Camera Communication(OCC). Rear RC Car follows the preceding one by analyzing transmitted OCC data based on image processing. Through this procedure, the information reception ratio according to the distance change of two RC Cars is confirmed. Through experiments, we showed that our proposed scheme enables the possibility of vehicle platooning.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.5
• /
• pp.313-322
• /
• 2020

A large volume of data will improve the robustness of deep learning models and avoid overfitting problems. In automatic tongue segmentation, the availability of annotated tongue images is often limited because of the difficulty of collecting and labeling the tongue image datasets in reality. Data augmentation can expand the training dataset and increase the diversity of training data by using label-preserving transformations without collecting new data. In this paper, augmented tongue image datasets were developed using seven augmentation techniques such as image cropping, rotation, flipping, color transformations. Performance of the data augmentation techniques were studied using state-of-the-art transfer learning models, for instance, InceptionV3, EfficientNet, ResNet, DenseNet and etc. Our results show that geometric transformations can lead to more performance gains than color transformations and the segmentation accuracy can be increased by 5% to 20% compared with no augmentation. Furthermore, a random linear combination of geometric and color transformations augmentation dataset gives the superior segmentation performance than all other datasets and results in a better accuracy of 94.98% with InceptionV3 models.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.6
• /
• pp.1183-1188
• /
• 2016

Kinect sensor version 2 is a kind of camera released by Microsoft as a computer vision and a natural user interface for game consoles like Xbox one. It allows acquiring color images, depth images, audio input and skeletal data with a high frame rate. In this paper, using depth image, we present a surveillance system of a certain area within Kinect's field of view. With computer vision library(Emgu CV), if an object is detected in the target area, it is tracked and kinect camera takes RGB image to send it in database server. Therefore, a mobile application on android platform was developed in order to notify the user that Kinect has sensed strange motion in the target region and display the RGB image of the scene. User gets the notification in real-time to react in the best way in the case of valuable things in monitored area or other cases related to a reserved zone.

• Korean Journal of Radiology
• /
• v.21 no.1
• /
• pp.58-67
• /
• 2020

Objective: Third-generation dual-source computed tomography (3rd-DSCT) allows dynamic myocardial CT perfusion imaging (dynamic CTP) with a 10.5-cm z-axis coverage. Although the increased radiation exposure associated with the 50% wider scan range compared to second-generation DSCT (2nd-DSCT) may be suppressed by using a tube voltage of 70 kV, it remains unclear whether image quality and the ability to quantify myocardial blood flow (MBF) can be maintained under these conditions. This study aimed to compare the image quality, estimated MBF, and radiation dose of dynamic CTP between 2ndDSCT and 3rd-DSCT and to evaluate whether a 10.5-cm coverage is suitable for dynamic CTP. Materials and Methods: We retrospectively analyzed 107 patients who underwent dynamic CTP using 2nd-DSCT at 80 kV (n = 54) or 3rd-DSCT at 70 kV (n = 53). Image quality, estimated MBF, radiation dose, and coverage of left ventricular (LV) myocardium were compared. Results: No significant differences were observed between 3rd-DSCT and 2nd-DSCT in contrast-to-noise ratio (37.4 ± 11.4 vs. 35.5 ± 11.2, p = 0.396). Effective radiation dose was lower with 3rd-DSCT (3.97 ± 0.92 mSv with a conversion factor of 0.017 mSv/mGy∙cm) compared to 2nd-DSCT (5.49 ± 1.36 mSv, p < 0.001). Incomplete coverage was more frequent with 2nd-DSCT than with 3rd-DSCT (1.9% [1/53] vs. 56% [30/54], p < 0.001). In propensity score-matched cohorts, MBF was comparable between 3rd-DSCT and 2nd-DSCT in non-ischemic (146.2 ± 26.5 vs. 157.5 ± 34.9 mL/min/100 g, p = 0.137) as well as ischemic myocardium (92.7 ± 21.1 vs. 90.9 ± 29.7 mL/min/100 g, p = 0.876). Conclusion: The radiation increase inherent to the widened z-axis coverage in 3rd-DSCT can be balanced by using a tube voltage of 70 kV without compromising image quality or MBF quantification. In dynamic CTP, a z-axis coverage of 10.5 cm is sufficient to achieve complete coverage of the LV myocardium in most patients.