• Journal of information and communication convergence engineering
• /
• v.10 no.2
• /
• pp.175-180
• /
• 2012

This paper presents a blind watermark detection scheme for an additive watermark embedding model. The proposed estimation-correlation-based watermark detector first estimates the embedded watermark by exploiting non-Gaussian of the real-world audio signal and the mutual independence between the host-signal and the embedded watermark and then a correlation-based detector is used to determine the presence or the absence of the watermark. For watermark estimation, blind source separation (BSS) based on independent component analysis (ICA) is used. Low watermark-to-signal ratio (WSR) is one of the limitations of blind detection with the additive embedding model. The proposed detector uses two-stage processing to improve the WSR at the blind detector; the first stage removes the audio spectrum from the watermarked audio signal using linear predictive (LP) filtering and the second stage uses the resulting residue from the LP filtering stage to estimate the embedded watermark using BSS based on ICA. Simulation results show that the proposed detector performs significantly better than existing estimation-correlationbased detection schemes.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.136-146
• /
• 2022

Some surface defects make the weak quality of steel materials. To limit these defects, we advocate a one-stage detector model RetinaNet among diverse detection algorithms in deep learning. There are several backbones in the RetinaNet model. We acknowledged two backbones, which are ResNet50 and VGG19. To validate our model, we compared and analyzed several traditional models, one-stage models like YOLO and SSD models and two-stage models like Faster-RCNN, EDDN, and Xception models, with simulations based on steel individual classes. We also performed the correlation of the time factor between one-stage and two-stage models. Comparative analysis shows that the proposed model achieves excellent results on the dataset of the Northeastern University surface defect detection dataset. We would like to work on different backbones to check the efficiency of the model for real world, increasing the datasets through augmentation and focus on improving our limitation.

• The Proceeding of the Korean Institute of Electromagnetic Engineering and Science
• /
• v.11 no.4
• /
• pp.39-52
• /
• 2000

A design of 2 stage S-DLVA(successive detector log video amplifier) was studied to detect wide dynamic radar pulse ranging from -70 ㏈m to 0㏈m. A basic design idea was focused on the linear detection in logarithmic scale of wide dynamic range radar pulses from nosie-like weak power of -70 ㏈m to relatively high power 0 ㏈m. It is highly formidable, since it requires high speed detection less than 10 nsec over the operating frequency ranges from 6 to 18 ㎓. A limiter diode, a tunnel diode and an L17-C were used as a protecting device, a detector diode and a log video amplifier in companion as a single stage detector to give voltage output proportional to the input power of about 35 ㏈ dynamic range. A protype of 2-stage DLVA having one more single stage detector was fabricated with a 32 ㏈ low noise amplifier and a 3 ㏈ hybrid coupler to provide total 70 ㏈ dynamic range detection. The logging characteristics were measured to have log slope of 25m.V/㏈ against 70 ㏈ logging range from -55 ㏈m to +15 ㏈m, the log linearity of within +/- 1.5 ㏈, and tangential sensitivity was at -63 ㏈m. The pulse dynamics of rise time and recovery time were measured as 50 nsec and 1.2 $\mu$sec, respectively. The reason might be due to the parasitic capacitances of packaged limiter, tunnel diode, and L17-C.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2002.09a
• /
• pp.318-321
• /
• 2002

As a part of the next generation PET project, we have developed a depth of interaction detector which is consist of three-dimensional arrays of GSO crystal elements sized 2.9mm ${\times}$ 2.9mm ${\times}$ 7.5mm. The basic structure of a detector block is 4-stages in depth, one stage is composed of 2 by 2 array of the crystal elements. The blocks are optically coupled to a position sensitive photomultiplier tube. Each crystal element can be in different conditions; rough or chemical etching for the crystal surface. The effect of the difference of crystal surface condition on the detector performance was analyzed in one-dimensional crystal array as a basic study for the three-dimensional detector by a simple model which is considered only probabilities of transmission, reflect and absorption of photons are in a crystal. As the next step, we investigated the effect of different crystal surface condition in a "U shaped detector" which is an array of stacked crystals bending at the center.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.6
• /
• pp.1-8
• /
• 1998

This paper presents an asynchronous DS/CDMA multiuser detector, which is a two stage, symbol-by-symbol scheme consisting of conventional detectors followed by linear decorrelating detectors. The conventional detector first makes temporal decisions and the detected symbols are delayed by one symbol period to be used for the selection of decorrelating bases in the subsequent decorrelaing detection stage. It also employs a bank of early-late correlators in place of a bank of single correlators taking the small offset of chip timing asynchronism into account. The proposed detector requires only the coarse knowledge of relative time delays of interfering users and is suitable for digital implementation. To verify the detector performance, the analytical BER performance will be given and compared with the simulation results for BPSK DS/CDMA signals in AWGN channel. While the performance of the proposed detector will be analyzed for time-limited signal, the simulation is carried out for both the time-limited and band-limited signals. As can be seen in the simulation results, the proposed scheme shows good results.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.427-430
• /
• 2019

Object detection is applied in various field. Autonomous driving, surveillance, OCR(optical character recognition) and aerial image etc. We will look at the algorithms that are using to object detect. These algorithms are divided into two methods. The one is R-CNN algorithms [2], [5], [6] which based on region proposal. The other is YOLO [7] and SSD [8] which are one stage object detector based on regression/classification.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.2
• /
• pp.1-9
• /
• 2020

A smart factory is a key industrial application introduced by the 4^th industrial revolution. The automatic guided vehicle (AGV) is one of the technology realizing smart factory, but the development cost is high due to its early stage of technology. Although developing a low-cost AGV requires a lot of data, it has limited data acquisition capability because of the limited storage and the AGV movement. Hence, we propose a development environment using Bluetooth to collect data and design a lane detector. The proposed lane detector shows a high lane detection ratio regardless of light variation and a shade.

• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.31-41
• /
• 2022

Steel is one of the most fundamental components to mechanical industry. However, the quality of products are greatly impacted by the surface defects in the steel. Thus, researchers pay attention to the need for surface defects detector and the deep learning methods are the current trend of object detector. There are still limitations and rooms for improvements, for example, related works focus on developing the models but don't take into account real-time application with practical implication on industrial settings. In this paper, a real-time application of steel surface defects detection based on YOLOv4 is proposed. Firstly, as the aim of this work to deploying model on real-time application, we studied related works on this field, particularly focusing on one-stage detector and YOLO algorithm, which is one of the most famous algorithm for real-time object detectors. Secondly, using pre-trained Yolov4-Darknet platform models and transfer learning, we trained and test on the hot rolled steel defects open-source dataset NEU-DET. In our study, we applied our application with 4 types of typical defects of a steel surface, namely patches, pitted surface, inclusion and scratches. Thirdly, we evaluated YOLOv4 trained model real-time performance to deploying our system with accuracy of 87.1 % mAP@0.5 and over 60 fps with GPU processing.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1993.10a
• /
• pp.243-248
• /
• 1993

New methodology for fiducial alignment is proposed to improve the alignment accuracy in wafer steppers. The positioning error is detected by PSD(Position Sensitive Detector)when 2-dimensional vernier patterns on a reticle on a reticle are projected on the fiducial marks of wafer stage. The width and period of vernier patterns are deter mined to get the highest S/N ratio for the exposure wavelength 248.4nm of KrF excimer laser. This new method has an advantage of higher accuracy and faster alignment over the conventional one.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1995.06a
• /
• pp.63-67
• /
• 1995

This paper proposes a new frequency offset correction technique for OFDM receivers on a frequency-selective fading channel. The frequency offset in the OFDM signals is known to introduce an interchannel interference among the multiple subcarriers, which degrades the receiver performance severely. In order to reduce the frequency offset, this paper describes an algorithm with two stages: acquisition and tracking. At both stages the algorithm oversamples the received OFDM signals. At the acquisition stage the frequency offset is reduced to half or less than the intercarrier spacing by matching the sign patterns of even and odd samples. Next, at tracking stage the frequency offset is compensated by a frequency detector which is controlled by the correlation of the even and odd sample sets. From the results, it is found that the proposed algorithm can correct the frequency offset even if the initial offset exceeds one half of th eintercairrers spacing.