• Journal of the Korea Society of Computer and Information
• /
• v.14 no.12
• /
• pp.63-73
• /
• 2009

This paper presents a vehicle detection method that uses convolution matching method based on a simple color information. An input image is expressed as 8 oriented color expression(Red, Green, Blue, White, Black, Cyan, Yellow, Magenta) considering an orientation of a pixel color vector. It makes the image very reliable and strong against changes of illumination condition or environment. This paper divides the vehicle detection into a hypothesis generation step and a hypothesis verification step. In the hypothesis generation step, the vehicle candidate region is found by vertical edge and shadow. In the hypothesis verification step, the convolution matching and the complexity of image edge are used to detect real vehicles. It is proved that the proposed method has the fast and high detection rate on various experiments where the illumination source and environment are changed.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.6
• /
• pp.759-764
• /
• 2005

In this paper we describe a face recognition using EBGM(Elastic Bunch Graph Matching) method. Usally, the PCA and LDA based face recognition method with the low-dimensional subspace representation use holistic image of faces, but this study uses local features such as a set of convolution coefficients for Gabor kernels of different orientations and frequencies at fiducial points including the eyes, nose and mouth. At pre-recognition step, all images are represented with same size face graphs and they are used to recognize a face comparing with each similarity for all images. The proposed algorithm has less computation time due to simplified face graph than conventional EBGM method and the fuzzy matching method for calculating the similarity of face graphs renders more face recognition results.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.138-141
• /
• 2004

본 논문은 EBGM(Elastic Bunch Graph Matching)기법을 이용한 얼굴인식에 대해 다룬다. 대용량 영상 정보에 대해 차원 축소를 이용한 얼굴인식 기법인 주성분기법이나 선형판별기법에서는 얼굴 영상 전체의 정보를 이용하는 반면 본 논문에서는 얼굴의 눈, 코, 입 등과 같은 얼굴 특징점에 대해 주파수와 방향각이 다른 여러 개의 가버 커널과 영상 이미지의 컨볼루션(Convolution)의 계수의 집합(Jets)을 이용한 특징 데이터를 이용한다. 하나의 얼굴 영상에 대해서는 모든 영상이 같은 크기의 특징 데이터로 표현되는 Face Graph가 생성되며, 얼굴인식 과정에서는 추출된 제트의 집합에 대해서 상호 유사도(Similarity)의 크기를 비교하여 얼굴인식을 수행한다. 본 논문에서는 기존의 EBGM방법의 Face Graph 생성 과정을 보다 간략화 한 방법을 이용하여 얼굴인식 과정에서 계산량을 줄여 속도를 개선하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.1C
• /
• pp.56-63
• /
• 2003

In this paper, a wavelet-based stereo matching algorithm to obtain an accurate disparity map in wavelet transformed domain by using a shift ability property, a modified wavelet transform, the similarities for their sub-bands, and a hierarchical structure is proposed. New approaches for stereo matching by lots of feature information are to utilize translation-variant results of the sub-bands in the wavelet transformed domain because they cannot literally expect translation invariance in a system based on convolution and sub-sampling. After the similarity matching for each sub-band, we can easily find optimal matched-points because the sub-bands appearance of the shifted signals is definitely different from that of the original signal with no shift.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.8 no.6
• /
• pp.620-628
• /
• 1997

When the Chebyshev function is applied to design the waveguide transition, it exhibits poor impedance matching characteristics due to impedance discontinuities at the ends of tapered line. In this paper, an improved Chebyshev function, which is obtained by using the convolution property, is proposed to make improvements on the impedance matching characteristics of the waveguide transition. When rectangular to circular waveguide transition is designed by improved function, then the computed return loss is approximately 5 dB better than the conventional Chebyshev function.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.5
• /
• pp.801-808
• /
• 2022

Research on how to embed knowledge in large-scale Linked Data and apply neural network models for entity matching is relatively scarce. The most fundamental problem with this is that different labels lead to lexical heterogeneity. In this paper, we propose an extended GCN (Graph Convolutional Network) model that combines re-align structure to solve this lexical heterogeneity problem. The proposed model improved the performance by 53% and 40%, respectively, compared to the existing embedded-based MTransE and BootEA models, and improved the performance by 5.1% compared to the GCN-based RDGCN model.

• The Journal of Korea Robotics Society
• /
• v.13 no.4
• /
• pp.230-236
• /
• 2018

3D depth perception has played an important role in robotics, and many sensory methods have also proposed for it. As a photodetector for 3D sensing, single photon avalanche diode (SPAD) is suggested due to sensitivity and accuracy. We have researched for applying a SPAD chip in our fusion system of time-of-fight (ToF) sensor and stereo camera. Our goal is to upsample of SPAD resolution using RGB stereo camera. Currently, we have 64 x 32 resolution SPAD ToF Sensor, even though there are higher resolution depth sensors such as Kinect V2 and Cube-Eye. This may be a weak point of our system, however we exploit this gap using a transition of idea. A convolution neural network (CNN) is designed to upsample our low resolution depth map using the data of the higher resolution depth as label data. Then, the upsampled depth data using CNN and stereo camera depth data are fused using semi-global matching (SGM) algorithm. We proposed simplified fusion method created for the embedded system.

• Journal of the Korean Society of Industry Convergence
• /
• v.17 no.3
• /
• pp.184-193
• /
• 2014

Machine vision systems is applied for various industries such as optimize your spending, automate your production and maximize your efficiency. This research is effective for most optimal light condition of machine vision that technology was applied bald outside human visual acuity. Image processing converts a target image captured by a CCD camera into a digital signal and then performs various arithmetic operations on the signal to extract the characteristics of the target, such as points, lines, circles, area and length. The mathematical concepts of convolution and the kernel matrix are used to apply filters to signals, to perform functions such as extracting edges and reducing unwanted noise. This research analyze and compares matching ratio with reference image and search for optimal lighting condition in accuracy that user wants coming input image according to brightness change of lighting.

• Journal of information and communication convergence engineering
• /
• v.11 no.4
• /
• pp.298-306
• /
• 2013

The scale invariant feature transform (SIFT) is an effective algorithm used in object recognition, panorama stitching, and image matching. However, due to its complexity, real-time processing is difficult to achieve with current software approaches. The increasing availability of parallel computers makes parallelizing these tasks an attractive approach. This paper proposes a novel parallel approach for SIFT algorithm implementation using a block filtering technique in a Gaussian convolution process on the SIMD Pixel Processor. This implementation fully exposes the available parallelism of the SIFT algorithm process and exploits the processing and input/output capabilities of the processor, which results in a system that can perform real-time image and video compression. We apply this implementation to images and measure the effectiveness of such an approach. Experimental simulation results indicate that the proposed method is capable of real-time applications, and the result of our parallel approach is outstanding in terms of the processing performance.