• Proceedings of the Optical Society of Korea Conference
• /
• 2004.02a
• /
• pp.130-131
• /
• 2004

• Proceedings of the Optical Society of Korea Conference
• /
• 2006.07a
• /
• pp.559-560
• /
• 2006

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.4
• /
• pp.288-294
• /
• 2017

Recently, deep learning using convolutional neural network (CNN) has been extensively studied in image recognition. Convolution consists of addition and multiplication. Multiplication is computationally expensive in hardware implementation, relative to addition. It is also important factor limiting a chip design in an embedded deep learning system. In this paper, I propose a parallel-addition processing algorithm that converts grayscale images to the superposition of binary images and performs convolution only with addition. It is confirmed that the convolution can be performed by a parallel-addition method capable of reducing the processing time in experiment for verifying the availability of proposed algorithm.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.4
• /
• pp.656-664
• /
• 2006

In this paper, we introduce a new LUT based watermarking method for a halftone image. Watermark bits are hidden at pseudo-random locations of halftone image in the proposed method. The pixel values of the halftone image are determined from the LUT entry indexed by both the neighborhood halftone pixels and current grayscale value. The LUT is trained by a set of grayscale images and corresponding halftone images. Advantage of the LUT method is that it can be executed very fast compared with other watermarking methods for a halftone image. Therefore, the algorithm can be embedded in a printer. Experiments for real scanned images showed that the method is a feasible method to hide the large amount of data within a halftone image without noticeable distortion and comparing to the DHED method, is almost same in quality but significantly shorten in processing time.

• Journal of Internet Computing and Services
• /
• v.21 no.5
• /
• pp.67-73
• /
• 2020

In this paper, we present a CNN-based gesture recognition approach which reduces the memory burden of input data. Most of the neural network-based gesture recognition methods have used a sequence of frame images as input data, which cause a memory burden problem. We use a motion history image in order to define a meaningful gesture. The motion history image is a grayscale image into which the temporal motion information is collapsed by synthesizing silhouette images of a user during the period of one meaningful gesture. In this paper, we first summarize the previous traditional approaches and neural network-based approaches for gesture recognition. Then we explain the data preprocessing procedure for making the motion history image and the neural network architecture with three convolution layers for recognizing the meaningful gestures. In the experiments, we trained five types of gestures, namely those for charging power, shooting left, shooting right, kicking left, and kicking right. The accuracy of gesture recognition was measured by adjusting the number of filters in each layer in the proposed network. We use a grayscale image with 240 × 320 resolution which defines one meaningful gesture and achieved a gesture recognition accuracy of 98.24%.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.4
• /
• pp.237-250
• /
• 2015

This paper describes a hardware-oriented algorithm and its conceptual implementation in a real-time speed limit traffic sign detection system on an automotive-oriented field-programmable gate array (FPGA). It solves the training and color dependence problems found in other research, which saw reduced recognition accuracy under unlearned conditions when color has changed. The algorithm is applicable to various platforms, such as color or grayscale cameras, high-resolution (4K) or low-resolution (VGA) cameras, and high-end or low-end FPGAs. It is also robust under various conditions, such as daytime, night time, and on rainy nights, and is adaptable to various countries' speed limit traffic sign systems. The speed limit traffic sign candidates on each grayscale video frame are detected through two simple computational stages using global luminosity and local pixel direction. Pipeline implementation using results-sharing on overlap, application of a RAM-based shift register, and optimization of scan window sizes results in a small but high-performance implementation. The proposed system matches the processing speed requirement for a 60 fps system. The speed limit traffic sign recognition system achieves better than 98% accuracy in detection and recognition, even under difficult conditions such as rainy nights, and is implementable on the low-end, low-cost Xilinx Zynq automotive Z7020 FPGA.

• Korean Journal of Optics and Photonics
• /
• v.11 no.1
• /
• pp.47-57
• /
• 2000

A two-stage iterative Fourier transform algorithm, based on hybrid input-output algorithm and new Pnoise algorithm, is used to design continuous and quantized phase-only diffractive pattern elements which produce arbitrary given intensity patterns via Fraunhofer diffraction. Numerical results for two $128\times128$ binary patterns and two grayscale patterns are compared with those of other algorithms. It is found that the algorithm yields better signal-to-noise ratio and even better uniformity with slightly lower diffraction efficiency than other algorithms. We investigated the dependence of performance on parameters used in the algorithm, size of noise region, and the number of phase levels for quantized elements. In the case of quantized phase elements, the size of noise region plays a greater role in determining the performance of the algorithm than given intensity pattern itself. tself.

• The KIPS Transactions:PartB
• /
• v.15B no.3
• /
• pp.229-236
• /
• 2008

This paper examines effectiveness of using color images instead of grayscale ones for automatic lipreading. First, we show the effect of color information for performance of humans' lipreading. Then, we compare the performance of automatic lipreading using features obtained by applying principal component analysis to grayscale and color images. From the experiments for various color representations, it is shown that color information is useful for improving performance of automatic lipreading; the best performance is obtained by using the RGB color components, where the average relative error reductions for clean and noisy conditions are 4.7% and 13.0%, respectively.

• Journal of Sensor Science and Technology
• /
• v.30 no.3
• /
• pp.170-174
• /
• 2021

As the rising attention to the medical and healthcare issue, Bio-MEMS (Micro electro mechanical systems) platform such as bio sensor, cell culture system, and microfluidics device has been studied extensively. Bio-MEMS platform mostly has high resolution structure made by biocompatible material such as polydimethylsiloxane (PDMS). In addition, three dimension structure has been applied to the bio-MEMS. Lithography can be used to fabricate complex structure by multiple process, however, non-rectangular cross section can be implemented by introducing optical apparatus to lithography technic. X-ray lithography can be used even for sub-micron scale. Here in, we demonstrated lines with round shape cross section using the tilted gold absorber which was deposited on the oblique structure as the X-ray mask. This structure was used as a mold for PDMS. Molded PDMS was applied to the cell culture platform. Moreover, molded PDMS was bonded to flat PDMS to utilize to the sub-micro channel. This work has potential to the large area bio-MEMS.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.905-907
• /
• 2022

악성코드를 분석하는 기존 기법인 정적분석은 빠르고 효율적으로 악성코드를 탐지할 수 있지만 난독화된 파일에 취약한 반면,, 동적분석은 난독화된 파일에 적합하지만 느리고 비용이 많이 든다는 단점을 가진다. 본 연구에서는 두 분석 기법의 단점을 해결하기 위해 딥러닝 모델을 활용한 난독화에 강한 정적분석 모델을 제안하였다. 본 연구에서 제안한 방법은 원본 코드 및 난독화된 파일을 grayscale 이미지로 변환하여 데이터셋을 구축하고 AutoEncoder 를 사전학습시켜 encoder 가 원본 파일과 난독화된 파일로부터 원본 파일의 특징을 추출할 수 있도록 한 이후, encoder 의 output 을 fully connected layer 의 입력으로 넣고 전이학습시켜 악성코드를 탐지하도록 하였다. 본 연구에서는 제안한 방법론은 난독화된 파일에서 악성코드를 탐지하는 성능을 F1 score 기준 14.17% 포인트 향상시켰고, 난독화된 파일과 원본 파일을 전체를 합친 데이터셋에서도 악성코드 탐지 성능을 F1 score 기준 7.22% 포인트 향상시켰다.