• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.1
• /
• pp.10-17
• /
• 2017

We propose classification algorithms for human and dog movement. The proposed algorithms use micro-Doppler signals obtained from humans and dogs moving in four different directions. A two-stage classifier based on a support vector machine (SVM) is proposed, which uses a radial-based function (RBF) kernel and $16^{th}$-order linear predictive code (LPC) coefficients as feature vectors. With the proposed algorithms, we obtain the best classification results when a first-level SVM classifies the type of movement, and then, a second-level SVM classifies the moving object. We obtain the correct classification probability 95.54% of the time, on average. Next, to deal with the difficult classification problem of human and dog running, we propose a two-layer convolutional neural network (CNN). The proposed CNN is composed of six ($6{\times}6$) convolution filters at the first and second layers, with ($5{\times}5$) max pooling for the first layer and ($2{\times}2$) max pooling for the second layer. The proposed CNN-based classifier adopts an auto regressive spectrogram as the feature image obtained from the $16^{th}$-order LPC vectors for a specific time duration. The proposed CNN exhibits 100% classification accuracy and outperforms the SVM-based classifier. These results show that the proposed classifiers can be used for human and dog classification systems and also for classification problems using data obtained from an ultra-wideband (UWB) sensor.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.6
• /
• pp.89-101
• /
• 2014

In this paper, we proposed a new local micro pattern, Signed Local Directional Pattern(SLDP). SLDP uses information of edges to represent the face's texture. This can produce a more discriminating and efficient code than other state-of-the-art methods. Each micro pattern of SLDP is encoded by sign and its major directions in which maximum edge responses exist-which allows it to distinguish among similar edge patterns that have different intensity transitions. In this paper, we divide the face image into several regions, each of which is used to calculate the distributions of the SLDP codes. Each distribution represents features of the region and these features are concatenated into a feature vector. We carried out facial expression recognition with feature vectors and SVM(Support Vector Machine) on Cohn-Kanade and JAFFE databases. SLDP shows better classification accuracy than other existing methods.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.2
• /
• pp.80-85
• /
• 2011

Alzheimer's disease and Parkinson's disease are the most common neurodegenerative disorders. In this paper, we proposed a feature extraction method for classification of AD and PD based on micro-Raman spectra from platelet. The first step of the preprocessing is a simple smoothing followed by background elimination to the original spectra to make it easy to measure the intensity of the peaks. The last step of the preprocessing was peak alignment with the reference peak. After the inspection of the preprocessed spectra, we found that proportion of two peak intensity at 743 and $757cm^{-1}$ and peak intensity at 1248 and $1448cm^{-1}$ are the most discriminative features. Then we apply mapstd method for normalization. The method returned data with means to 0 and deviation to 1. With these three features, the classification result involving 263 spectra showed about 95.8% true classification in case of MAP(maximum a posteriori probability).

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.49 no.4
• /
• pp.7-14
• /
• 2012

A image quality is limited by a sample-and-hold circuit of the X-ray CMOS image sensor even though simple mos switch or bootstrapped clock circuit are used to get high quality sampled signal. Because distortion of sampled signal is produced by the charge injection from sample-and-hold circuit even using bootstrapped. This paper presents the 3D micro-inductor design methode in the CMOS process. Using this methode, it is possible to increase the ENOB (effective number of bit) through the use of micro-inductor which is calculated and designed in standard CMOS process in this paper. The ENOB is improved 0.7 bit from 17.64 bit to 18.34 bit without any circuit just by optimized inductor value resulting in verified simulation result. Because of this feature, micro-inductor methode suggested in this paper is able to adapt a mamography that is needed high resolution so that it help to decrease patients dose amount.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.12
• /
• pp.1875-1882
• /
• 2013

Micro-array technology contributes numerous achievements such as ordering of gene network and integration of genomic. This technology is well established as means for investigating patterns of gene expression. DNA micro-arrays utilize Affymetric chips where a large quantity of DNA sequences may be synthesized. There are two general type of conventional DNA array spotter: contact and piezoelectric. The contact technology used spotting pin technology to make contact with the glass slide surface. This may caused damage or scratches to the surface matrix where protein will be contaminated and may not bind specifically. Piezoelectric technology available at this present time on the other hand requires the analyzer to print the result that can only be done within the laboratory despite of mass production. Therefore, in this paper, high-throughput technology is developed for providing greater consistency in feature spot without touching the glass slide surface.

• Transactions on Electrical and Electronic Materials
• /
• v.3 no.1
• /
• pp.30-37
• /
• 2002

A focusing grating coupler (FGC) was not fabricated by the 'Continuous Path Control'writing strategy but by an electron-beam lithography system of more general exposure mode, which matches not only the address grid with the grating period but also an integer multiple of the address grid resolution (5 nm). To more simplify the fabrication, we are able to reduce a process step without large decrease of pattern quality by excluding a conducting material or layer such as metal (Al, Cr, Au), which are deposited on top or bottom of an e-beam resist to prevent charge build-up during e-beam exposure. A grating pitch period and an aperture feature size of the FGC designed and fabricated by e-beam lithography and reactive ion etching were ranged over 384.3 nm to 448.2 nm, and 0.5 $\times$ 0.5 mm$^2$area, respectively. This fabrication method presented will reduce processing time and improve the grating quality by means of a consideration of the address grid resolution, grating direction, pitch size and shapes when exposing. Here our investigations concentrate on the design and efficient fabrication results of the FGC for coupling from slab waveguide to a spot in free space.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.1
• /
• pp.142-148
• /
• 2003

Reinforcement learning is considered as an important tool for robotic learning in unknown/uncertain environments. In this paper, we propose an evaluation function expressed in a vector form to realize multi-dimensional reinforcement learning. The novel feature of the proposed method is that learning one behavior induces parallel learning of other behaviors though the objectives of each behavior are different. In brief, all behaviors watch other behaviors from a critical point of view. Therefore, in the proposed method, there is cross-criticism and parallel learning that make the multi-dimensional learning process more efficient. By ap-plying the proposed learning method, we carried out multi-dimensional evaluation (reward) and multi-dimensional learning simultaneously in one trial. A special neural network (Q-net), in which the weights and the output are represented by vectors, is proposed to realize a critic net-work for Q-learning. The proposed learning method is applied for behavior planning of mobile robots.

• Design & Manufacturing
• /
• v.2 no.5
• /
• pp.5-10
• /
• 2008

Recently, ultra precision and light-weight micro products are needed in various industries. Injection molding products with metal insert material is often satisfied with light-weight and precision simultaneously. The researches on macro-size insert deformation have been performed but, a research on micro-size insert is meager. In this paper, the injection molding product with $300{\mu}m$ thin foil insert is designed and insert injection molding process is performed. Finally, the deformation of thin foil insert is analyzed according to insert feature and gate length.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.464-467
• /
• 2005

Among several polishing techniques for micro structures, polishing process using magnetorheological(MR) fluid has advantages in the finishing process of 3-D micro structures because abrasives in the fluid can reach surfaces with complex feature and play their role. Although many researchers have been trying to reveal its polishing mechanism of the MR polishing, it has not been successful because in-situ measurement of state variables is difficult and process parameters are complex. In fact, one of the key factors for applying process control methodologies, such as Run-to-Run control, is the measuring and monitoring of slurry quality because the process strongly depends on the fluid property. Therefore, it is necessary to maintain consistent slurry quality to guarantee the process repeatability. The proposed equipment achieves the longer life cycle of MR fluid and reduces the variability of products. A new method to measure the material removal rate in MRF polishing process is also proposed and discussed.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1871-1874
• /
• 2005

The direct write FIB technology has several advantages over contemporary micro-machining technology, including better feature resolution with low lateral scattering and capability of maskless fabrication. Therefore, the application of FIB technology in micro fabrication has become increasingly popular. In recent model of FIB, however the feeding system has been a very coarse resolution of about a few ${\mu}m$. Our research is the development of nano stage of 200mm strokes and 10nm resolutions. Also, this stage should be effectively operating in ultra high vacuum of about $1x10^{-7}$ torr. This paper presents the discussion and results of CAE of the 2 axes stages. we have estimated the stable static and dynamic characteristics for dual servo system. Therefore the 2 axes stages developed and future work are introduced at the end of the paper.