• Journal of the Institute of Electronics Engineers of Korea SC
• /
• 제46권5호
• /
• pp.29-37
• /
• 2009

This paper describes a vision-based and real-time system for detecting road signs from within a moving vehicle. The proposed system has the standard architecture with adaboost algorithm to detect road signs in real time. And it uses the value of vortical plane in the process of extracting candidate areas in view of fact that there are vertically most of signs on roads. Although being useful for detecting objects in real time, the conventional adaboost algorithm deteriorates the performance of detection rate in complex circumstance by reason of using only integral images as features. To overcome this problem, this paper proposes the method that improves the reliability of candidates as using the value of vertical plane for extracting candidate area and improves the performance of the detection rate as using integral images to which we add the kind of feature prototype. The experiments of this paper show that the detection rate of the proposed method has higher than that of the conventional adaboost algorithm under the real complex circumstance of roads.

• Journal of Korea Multimedia Society
• /
• 제13권9호
• /
• pp.1314-1321
• /
• 2010

In this paper, we study about the vehicle detection algorithm which is in the process of travelling from the road. An input image is segmented by means of split and merge algorithm. And two largest segmented regions are removed for reducing search region and speed up processing time. In order to detect the back side of the front vehicle considers a vertical/horizontal component, uses an integral image with to apply Haar-like methods which are the possibility of shortening a calculation time, classified with SVM. The simulation result of the method which is proposed appeared highly.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제17권1호
• /
• pp.110-119
• /
• 2017

This paper describes a CMOS image sensor (CIS) with dual correlated double sampling (CDS) and column-parallel analog-to-digital converter (ADC) and its measurement method using a field-programmable gate array (FPGA) integrated module. The CIS is composed of a $320{\times}240$ pixel array with $3.2{\mu}m{\times}3.2{\mu}m$ pixels and column-parallel 10-bit single-slope ADCs. It is fabricated in a $0.11-{\mu}m$ CIS process, and consumes 49.2 mW from 1.5 V and 3.3 V power supplies while operating at 6.25 MHz. The measured dynamic range is 53.72 dB, and the total and column fixed pattern noise in a dark condition are 0.10% and 0.029%. The maximum integral nonlinearity and the differential nonlinearity of the ADC are +1.15 / -1.74 LSB and +0.63 / -0.56 LSB, respectively.

• Journal of Institute of Control, Robotics and Systems
• /
• 제20권9호
• /
• pp.930-935
• /
• 2014

A Gabor filter is a linear filter used for edge detectionas frequency and orientation representations of Gabor filters are similar to those of the human visual system. In this thesis, we propose a pedestrian detection algorithm using a Gabor filter bank. In order to extract the features of the pedestrian, we use various image processing algorithms and data structure algorithms. First, color image segmentation is performed to consider the information of the RGB color space. Second, histogram equalization is performed to enhance the brightness of the input images. Third, convolution is performed between a Gabor filter bank and the enhanced images. Fourth, statistical values are calculated by using the integral image (summed area table) method. The calculated statistical values are used for the feature matrix of the pedestrian area. To evaluate the proposed algorithm, the INRIA pedestrian database and SVM (Support Vector Machine) are used, and we compare the proposed algorithm and the HOG (Histogram of Oriented Gradient) pedestrian detector, presentlyreferred to as the methodology of pedestrian detection algorithm. The experimental results show that the proposed algorithm is more accurate compared to the HOG pedestrian detector.

• The Transactions of the Korean Institute of Power Electronics
• /
• 제8권5호
• /
• pp.418-426
• /
• 2003

To show a retina shape and thickness on the computer monitor, a laser has been used in Scanning Laser Ophthalmoscope(SLO) equipment using the traveling difference. This method requires exact synchronous control of laser traveling in optic system to show a clear 3-dimensional image of retina To obtain this image, this exact synchronism is very important for making the perfect plane scanning. In this study, a high speed and synchronous control of the galvanometer to make 3-dimensional retina image is presented. For the more, deadbeat load torque observer is added to the PI controller for compensation of the position error arisen in the high speed control. As a result, the proposed control system has a robust and precise response against the load torque variation appeared in high speed control. A stability and usefulness are verified by the computer simulation and the experiment.

• Journal of the Semiconductor & Display Technology
• /
• 제11권1호
• /
• pp.79-83
• /
• 2012

We use adaptive method and determine threshold coefficient so that the algorithm could decide a suitable binarization threshold coefficient of the image to detecting a marker; therefore, we solve the light influence on the shadow area and dark region. In order to improve the speed for reducing computation we created Integral Image. The algorithm detects an outline of the image by using canny edge detection for getting damage or obscured markers as it receives the noise removed picture. The strength of the line of the outline is extracted by Hough transform and it extracts the candidate regions corresponding to the coordinates of the corners. Markers extracted using the equation of a straight edge to find the coordinates. By using the equation of straight the algorithm finds the coordinates the corners. of extracted markers. As a result, even if all corners are obscured, the algorithm can find all of them and this was proved through the experiment.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 한국정보통신학회 2013년도 추계학술대회
• /
• pp.77-80
• /
• 2013

SURF algorithm is an algorithm to extract feature points and to generate descriptors from input images. It is robust to change of environment such as scale, rotation, illumination and view points. Because of these features, it is used for many image processing applications such as object recognition, constructing panorama pictures and 3D image restoration. But there is disadvantage for real time operation because many recognition algorithms such as SURF algorithm requires a lot of calculations. In this paper, we propose a design of feature extractor and descriptor generator based on SURF for high memory efficiency. The proposed design reduced a memory access and memory usage to operate in real time.

• ETRI Journal
• /
• 제31권6호
• /
• pp.717-724
• /
• 2009

This paper describes a 1.2 V 12 b 60 MS/s CMOS analog front-end (AFE) employing low-power and flexible design techniques for image signal processing. An op-amp preset technique and programmable capacitor array scheme are used in a variable gain amplifier to reduce the power consumption with a small area of the AFE. A pipelined analog-to-digital converter with variable resolution and a clock detector provide operation flexibility with regard to resolution and speed. The AFE is fabricated in a 0.13 ${\mu}m$ CMOS process and shows a gain error of 0.68 LSB with 0.0352 dB gain steps and a differential/integral nonlinearity of 0.64/1.58 LSB. The signal-to-noise ratio of the AFE is 59.7 dB at a 60 MHz sampling frequency. The AFE occupies 1.73 $mm^2$ and dissipates 64 mW from a 1.2 V supply. Also, the performance of the proposed AFE is demonstrated by an implementation of an image signal processing platform for digital camcorders.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권4호
• /
• pp.1166-1186
• /
• 2022

Artistic font design has become an integral part of visual media. However, without prior knowledge of the font domain, it is difficult to create distinct font styles. When the number of characters is limited, this task becomes easier (e.g., only Latin characters). However, designing CJK (Chinese, Japanese, and Korean) characters presents a challenge due to the large number of character sets and complexity of the glyph components in these languages. Numerous studies have been conducted on automating the font design process using generative adversarial networks (GANs). Existing methods rely heavily on reference fonts and perform font style conversions between different fonts. Additionally, rather than capturing style information for a target font via multiple style images, most methods do so via a single font image. In this paper, we propose a network architecture for generating multilingual font sets that makes use of geometric structures as content. Additionally, to acquire sufficient style information, we employ multiple style images belonging to a single font style simultaneously to extract global font style-specific information. By utilizing the geometric structural information of content and a few stylized images, our model can generate an entire font set while maintaining the style. Extensive experiments were conducted to demonstrate the proposed model's superiority over several baseline methods. Additionally, we conducted ablation studies to validate our proposed network architecture.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• 제47권1호
• /
• pp.160-170
• /
• 2010

This paper proposes a Haar Wavelet Feature Detector (HWFD) based on the Haar wavelet transform and average box filter. By decomposing the original image using the Haar wavelet transform, the proposed detector obtains the variance information of the image, making it possible to extract more distinctive features from the original image. For detection of interest points that represent the regions whose variance is the highest among their neighbor regions, we apply the average box filter to evaluate the local variance information and use the integral image technique for fast computation. Due to utilization of the Haar wavelet transform and the average box filter, the proposed detector is robust to illumination change, scale change, and rotation of the image. Experimental results show that even though the proposed method detects fewer interest points, it achieves higher repeatability, higher efficiency and higher matching accuracy compared with the DoG detector and Harris corner detector.