• Journal of Korea Multimedia Society
• /
• v.24 no.4
• /
• pp.528-537
• /
• 2021

Face recognition has gained significant notice because of its application in many businesses: security, healthcare, and marketing. In this paper, we will present the recognition method using the combination of correlation filters (CF) and Support Vector Machine (SVM). Firstly, we evaluate the performance and compared four different correlation filters: minimum average correlation energy (MACE), maximum average correlation height (MACH), unconstrained minimum average correlation energy (UMACE), and optimal-tradeoff (OT). Secondly, we propose the machine learning approach by using the OT correlation filter for features extraction and SVM for classification. The numerical results on National Cheng Kung University (NCKU) and Pointing'04 face database show that the proposed method OT-SVM gets higher accuracy in face recognition compared to other machine learning methods. Our approach doesn't require graphics card to train the image. As a result, it could run well on a low hardware system like an embedded system.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.29A no.7
• /
• pp.48-55
• /
• 1992

In this paper, fSDF/POF, BPOF spatial matched filters are designed and implemented by CGH. The correlation characteristics for the suggested filters are analyzed for the distorted input images. Input patterns are obtained from the out-of-plane aircraft images by gradually rotating it, and then used for SDF training images. Modified version of LCD is used for a real-time input device of an optical correlator, and CGH-based fSDF filters are fabricated on film mask for spatial matched filter in order to recognize the distorted images. Total optical corrlator system size is effectively reduced to 148.8 cm by using lens combinations. Computer simulations and experimental results show that the suggested phase filters have nearly uniform correlation characteristics and have good classification capabilities between two classes.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.4
• /
• pp.44-55
• /
• 1993

In this paper, we proposed a new optical method for the efficient recognition of Korean characters. There are six filters in the proposed method which employed the concepts of amplitude-modulated phase-only filter(AMPOF) and spatial frequency modulation(SFM). Here, amplitude modulation is used to achieve improved correlation discrimination and SFM is to reduce the number of filters. We also used a simplified synthetic discriminant function(SDF) for distortion invariance of input image. In order to recognize the partial rotation invariant Korean characters, the proposed distortion invariant minimum average correlation energy (MACE) filter is synthesized SFM, partial rotation invariant filter (PRIF), AMPOF and MACE for partial rotation invariance in the frequency domain. The advantage of the proposed filters is to supress the sidelobes of cross correlation peak away from the autocorrelation peak and to produce sharp correlation peaks. We performed simulation and optical experiment for some of Korea characters using the proposed method. The results show that the proposed method has more improved discriminant ability and reduced processing time than the conventional methods.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.6
• /
• pp.277-283
• /
• 2021

This paper addresses the problem of vision-based mask filter inspection for mask production systems. Machine learning-based approaches can be considered to solve the problem, but they may not be applicable to mask filter inspection if normal and anomaly mask filter data are not sufficient. In such cases, handcrafted image processing methods have to be considered to solve the problem. In this paper, we propose a hierarchical correlation-based approach that combines handcrafted image processing methods to detect anomaly mask filters. The proposed approach combines image rotation, cropping and resizing, edge detection of mask filter parts, average blurring, and correlation-based decision. The proposed approach was tested and analyzed with real mask filters. The results showed that the proposed approach was able to successfully detect anomalies in mask filters.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.2
• /
• pp.392-400
• /
• 1997

In this paper we porposed WCHF-fSDF filter to obtain a roration and scale-invariant correlation output. WCHF-fSDF filter is synthesized by each single CHF exttracted from scale-changed and wavelet tranformed imagesfor a refereence image as tranining images. The wavelet transform is defined as the correlation of an input image with a wavelet function. Therefore two 4f optical correlation systems are needed for pattern recognition using wavelet transform. We here include the wavelet function for the input image in the process of the proposed filter design and substitute the two 4f optical correlation system with a single 4f optical correlation system. The Performances of the proposed filter are compared with conventional CHF-SDF, POCHF-SDF filters through the computer simulation. The results of computer simulation show that the proposed filter has the rotation and scale-invariant correlation output and it has better performances than thoseof the conventioanl filters.

• Biomolecules & Therapeutics
• /
• v.10 no.4
• /
• pp.246-252
• /
• 2002

In recent years, the safety of sunscreens has been challenged based on the reports of their adverse effect on users; dermatitis, allergic contact dermatitis and photo allergic contact dermatitis. The unscientific idea that higher SPF sunsreen is good for health mealeads many users to tend to prefer higher SPF sunscreen. In the toxicological aspect, the need to investigate the safety of sunscreens is steadily increasing. However, there were few studies on the correlation between sun protection factor (SPF) and the safety of sunscreens. The objective of this study was to assess whether there was a correlation between SPF and the safety of sunscreens. We measured in vitro SPF index using homosalate as a standard and examined the toxicity tests such as primary skin irritation tests, ocular irritation test and skin sensitization test. Homosalate (HS), octyl methoxycinnamate (OMC), octyl salicylate (OS), octocrylene (OC) as UVB organic filter and benzophenone-3 (BP3), butyl methoxy dibenzoil methane (BMDM) as UVA organic filter, and titanium dioxide (TD), zinc oxide (ZO) as inorganic filters were used. The skin irritation indexes in rabbits treated with HS, OMC, OS, BP3, and BMDM were significantly increased as SPF indices were increased. Neither ocular irritation in rabbits nor skin sensitization in guinea pigs were increased. It suggests that there might be a good correlation between SPF and the skin irritation indices of organic UV filters and skin irritation might be one of most sensitive index to assess the safety of sunscreens.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.6
• /
• pp.899-908
• /
• 1995

In this paper, the probabilistic relations of dual morphological operations, such as dilation and erosion, closing and opening, and close- open and open-close, and the statistical properties for a continuous stationary lst order Gauss-Markov source are analyzed. The result is that the dual filters have symmetrical means and skews, and equal variances. Also, the statistics of morphological filters are very similar with those of input source, as correlation coefficient increases.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.3
• /
• pp.109-119
• /
• 1996

We propsoed a multiplexed optical correlation filter method for an optical parallel addition based on symbolic substitution. In the proposed mthod, we used redundant binary number which was easy to minimize the number of the symbolic substitution rules. We chose MACE filter which had very low sidelobes and good correlation peak compared with SDF filter as the optical correlation recognition filter and encoded input numbers properly to increase the discrimination capability. In order to minimize the number of symbolic substitution rules, sixteen input patterns were divided into six groups of the same addition results and six filters for recognizing the input patterns were used. these filters were multiplexed in two MMACE filter planes and the corresponding substitution method was proposed. Through the computer simulation, we confirmed the proposed method was suitable to implement the optical parallel adder.

• The Bulletin of The Korean Astronomical Society
• /
• v.42 no.2
• /
• pp.89.1-89.1
• /
• 2017

In this poster, we present standard calibration processes and results for photometry of $H-{\alpha}$ filter using broadband filters. We took data from a night in stable weather condition. And we derived parameters for band pass calibration including color terms. The corrected photometry results from broadband filters like V and R filters showed high correlation enough to replace instrumental $H-{\alpha}$ magnitude. We plan to extend these standardizing processes to another narrowband filters and flux calibration of narrowband filters from photometry of PNe.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.4
• /
• pp.701-711
• /
• 1994

In this paper, we design wavelet matched filters using Wavelet Transform available in the field of time-frequency analysis, for the optical patter recognition, and study optical correlation properites according to variations of the dilation factor of wavelet functions, which were compared with conventional matched filter and POF. We have found that the wavelet matched filters depend upon the shapes if used input patterns and dilation factors of the wavelet function in order to have the best dicriminability. Also, we show the edge enhanced effects of 2-D image, which is one of the properities of the wavelet transform. If has been found that the wavelet matched filter can discriminate better than the conventional matched filter and POF owing to their bandpass properties, and found that the proposed wavelet matched filters have more sensitive correlation response than POF.