• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.9C
• /
• pp.748-755
• /
• 2010

In this paper we propose an improved face recognition method using Gabor tensor. Gabor transform is known to be able to represent characteristic feature in face and reduced environmental influence. It may contribute to improve face recognition ratio. We attempted to combine three-dimensional tensor from Gabor transform with MPCA(Multilinear PCA) and LDA. MPCA with tensor which use various features is more effective than traditional one or two dimensional PCA. It is known to be robust to the change of face expression or light. Proposed method is simulated by MATALB9 using ORL and Yale face database. Test result shows that recognition ratio is improved maximum 9~27% compared with exisisting face recognition method.

• Smart Media Journal
• /
• v.9 no.2
• /
• pp.92-98
• /
• 2020

This paper proposes a vehicle license plate recognition system using light weight deep learning models without high-end server. The proposed license plate recognition system consists of 3 steps: [license plate detection]-[character area segmentation]-[character recognition]. SSD-Mobilenet was used for license plate detection, ResNet with localization was used for character area segmentation, ResNet was used for character recognition. Experiemnts using Samsung Galaxy S7 and LG Q9, accuracy showed 85.3% accuracy and around 1.1 second running time.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.11
• /
• pp.1527-1535
• /
• 1995

This paper introduces an isolated word recognition system realized on a neurocomputer called E-MIND II, which is a 2-D torus wavefront array processor consisting of 256 DNP IIs. The DNP II is an all digital VLSI unit processor for the EMIND II featuring the emulation capability of more than thousands of neurons, the 40 MHz clock speed, and the on-chip learning. Built by these PEs in 2-D toroidal mesh architecture, the E- MIND II can be accelerated over 2 Gcps computation speed. In this light, the advantages of the E-MIND II in its capability of computing speed, scalability, computer interface, and learning are especially suitable for real time application such as speech recognition. We show how to map a TDNN structure on this array and how to code the learning and recognition algorithms for a user independent isolated word recognition. Through hardware simulation, we show that recognition rate of this system is about 97% for 30 command words for a robot control.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.1
• /
• pp.13-21
• /
• 2018

AGVS (Automated Guided Vehicle System) is a core technology of logistics automation which automatically moves specific objects or goods within a certain work space. Conventional AGVS generally requires the in-door localization system and each AGV equips expensive sensors such as laser, magnetic, inertial sensors for the route recognition and automatic navigation. thus the high installation cost is inevitable and there are many restrictions on route(path) modification or expansion. To address this issue, in this paper, we propose a cost-effective and scalable AGV based on a light-weight pattern recognition technique. The proposed pattern recognition technology not only enables autonomous driving by recognizing the route(path), but also provides a technique for figuring out the loc ation of AGV itself by recognizing the simple patterns(bar-code like) installed on the route. This significantly reduces the cost of implementing AGVS as well as benefiting from route modification and expansion. In order to verify the effectiveness of the proposed technique, we first implement a pattern recognition algorithm on a light-weight MCU(Micro Control Unit), and then verify the results by implementing an MCU_controlled AGV prototype.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2009.11a
• /
• pp.284-284
• /
• 2009

The LED of cultivation edible plants was compliance the variable of the photo-receptor pigment with the red light source and ultra red light source from long wave region. The mechanism of cultivation edible plants for each part was necessary the wavelength unit which is appropriate, the illuminant source, motor control and lens design of LED light source about plant. The photo-receptor pigment induces for a long daytime recognition, seed germination and anthesis etc, induction years exists in the state which is an inactivity within the cells and in compliance with the red light source to be converted in active

• BMB Reports
• /
• v.38 no.6
• /
• pp.690-694
• /
• 2005

Transcription termination of the human mitochondrial genome requires specific binding to termination factor mTERF. In this study, mTERF was produced in E. coli and purified by two-step chromatography. mTERF-binding DNA sequences were isolated from a pool of randomized sequences by the repeated selection of bound sequences by gel-mobility shift assay and polymerase chain reaction. Sequencing and comparison of the 23 isolated clones revealed a 16-bp consensus sequence of 5'-GTG$\b{TGGC}$AGANCCNGG-3' in the light-strand (underlined residues were absolutely conserved), which nicely matched the genomic 13-bp terminator sequence 5'-$\b{TGGC}$AGAGCCCGG-3'. Moreover, mTERF binding assays of heteroduplex and single-stranded DNAs showed mTERF recognized the light strand in preference to the heavy strand. The preferential binding of mTERF with the light-strand may explain its distinct orientation-dependent termination activity.

• Journal of Magnetics
• /
• v.8 no.1
• /
• pp.50-59
• /
• 2003

Spatial light modulators (SLMs) are centrally important devices in volumetric recording, data Processing, Pattern recognition and other optical systems. Various types of reusable SLMs with two-dimensional pixel arrays have been intensively developed. Of these, magneto-optic spatial light modulators (MOSLMs) have advantages of high switching speed, robustness, nonvolatility, and radioactive resistance. In this article, we review recent development work on MOSLMs, mainly in relation to our own studies.

• International Journal of Quality Innovation
• /
• v.4 no.2
• /
• pp.50-64
• /
• 2003

During IC inspection, which includes the two parts of Mark and Lead, the deviation of IC on the tape occurring in high speed movements usually generates light reflection effect, which in turn causes errors in IC recognition as measured by machine vision system. This research filters the light reflection effect by developing standard components, identifies the correct position of IC Lead, hence fixes the measurement errors or non-measurability caused by light reflection, avoids the resulting discontinued operation of measuring system, and improves the productivity.

• Current Optics and Photonics
• /
• v.5 no.5
• /
• pp.506-513
• /
• 2021

An arrayed diffractive optical element design for the beam-shaping of a multi-array light source is proposed. This is an essential device for recent optical security and face recognition applications. In practice, we devise a DC noise reduction technique featuring high fabrication error tolerance regarding the multi-array light source diffractive optical elements, as a necessary part of the proposed design method. The spherical diverging illumination leads to DC-conjugate noise spreading. The main idea is tested experimentally, and the multi-array light source diffraction pattern is investigated numerically.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.991-998
• /
• 2022

Recently, object recognition using image/video signals is rapidly spreading on autonomous driving and mobile phones. However, the actual input image/video signals are easily exposed to a poor illuminance environment. A recent researches for improving illumination enable to estimate and compensate the illumination parameters. In this study, we propose VE-DCE (video enhancement zero-reference deep curve estimation) to improve the illumination of low-light images. The proposed VE-DCE uses unsupervised learning-based zero-reference deep curve, which is one of the latest among learning based estimation techniques. Experimental results show that the proposed method can achieve the quality of low-light video as well as images compared to the previous method. In addition, it can reduce the computational complexity with respect to the existing method.