• 한국멀티미디어학회논문지
• /
• 제22권7호
• /
• pp.790-803
• /
• 2019

The effects of different heat treatments on the sensory characteristics of abalones are studied in this study. In this paper, the sensory evaluation of abalone samples under different heat treatment conditions is carried out, and the evaluation results are analyzed. The three-dimensional (3D) scanning and reverse engineering are used in tooth modeling of the sensory evaluation of abalone samples under different heat treatment conditions. Besides, the chewing movement models are simplified into three modes, including the cutting mode, compressing mode and grinding mode, which are simulated using finite element simulation. The elastic modulus of the abalone samples is obtained through the compression testing using a texture analyzer to distinguish their material properties under different heat treatments and to obtain simulated mechanical parameters. Finally, taking the mechanical parameters of the finite element simulation of abalone chewing as input and sensory evaluation parameters as the output, BP neural network is established in which the sensory texture evaluation model of abalone samples is obtained. Through verification, the neural network prediction model can meet the requirements of food texture evaluation, with an average error of 9.12%.

• Journal of Biosystems Engineering
• /
• 제33권5호
• /
• pp.309-316
• /
• 2008

The goal of this study was to develop prediction models to estimate the storage days of tomato. The transmittance spectral data measured on tomato were preprocessed through normalization, SNV, Savitzky-Golay, and Norris Gap and then were used to build the prediction models using partial least square (PLS) method. For the experiments, the tomato samples of different varieties were collected at different harvest time. The samples were taken right after harvest from the field and then were stored in a low-temperature storage room in which room temperature was maintained at $10^{\circ}C$. The transmittance spectral data of the tomato samples were measured at three-day intervals for 16 days. The performance of the prediction models was affected by the preprocessing techniques as well as the varieties and harvest time of the tomato. The best model was found when SNV was applied. The accuracy of the best model was 90.2%. It can be concluded that the transmittance spectra are useful information for predicting the period of storage of tomato.

• Food Science and Biotechnology
• /
• 제18권2호
• /
• pp.413-418
• /
• 2009

Effects of chlorophyll and visible light exposure on the volatile formations and headspace oxygen content were studied in lard model systems at $55^{\circ}C$. Samples with or without addition of chlorophyll under light underwent photosensization or photooxidation, respectively. Total volatiles (TI) in lard with 5 ppm chlorophyll photosensization were 19 times higher than those in visible light photooxidized samples for 48 hr while TI in lard with chlorophyll in the dark were not significantly different from those in photooxidized samples (p>0.05). Headspace oxygen content in photosensitized lard decreased from 21 to 15% for 48 hr but that in photooxidized lard or that in lard with chlorophyll in the dark did not change significantly (p>0.05), which indicates that lard system used in this study is a photosensitizer-free model system and the presence of chlorophyll accelerated the lipid oxidation only under visible light. Oxidation mechanisms of photooxidation with or without presence of photosensitizers under visible light were not the same based on the difference of oxidized volatile profiles and headspace oxygen depletion.

• 한국축산식품학회지
• /
• 제36권6호
• /
• pp.744-751
• /
• 2016

The objective of this study was to investigate the effects of partial beef fat replacement (0, 30, 50, 100%) with gelled emulsion (GE) prepared with olive oil on functional and quality properties of model system meat emulsion (MSME). GE consisted of inulin and gelatin as gelling agent and characteristics of gelled and model system meat emulsions were investigated. GE showed good initial stability against centrifugation forces and thermal stability at different temperatures. GE addition decreased the pH with respect to increase in GE concentration. Addition of GE increased lightness and yellowness but reduced redness compared to control samples. The results of the study showed that partial replacement of beef fat with GE could be used for improving cooking yield without negative effects on water holding capacity and emulsion stability compared to C samples when replacement level is up to 50%. The presence of GE significantly affected textural behaviors of samples (p<0.05). In conclusion, our study showed that GE have promising impacts on developing healthier meat product formulations besides improving technological characteristics.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제14권1호
• /
• pp.96-99
• /
• 2014

In high-${\kappa}$/metal-gate (HK/MG) metal-oxide-semiconductor field-effect transistors (MOSFETs) at 45-nm and below, the metal-gate material consists of a number of grains with different grain orientations. Thus, Monte Carlo (MC) simulation of the threshold voltage ($V_{TH}$) variation caused by the workfunction variation (WFV) using a limited number of samples (i.e., approximately a few hundreds of samples) would be misleading. It is ideal to run the MC simulation using a statistically significant number of samples (>~$10^6$); however, it is expensive in terms of the computing requirement for reasonably estimating the WFV-induced $V_{TH}$ variation in the HK/MG MOSFETs. In this work, a simple matrix model is suggested to implement a computing-inexpensive approach to estimate the WFV-induced $V_{TH}$ variation. The suggested model has been verified by experimental data, and the amount of WFV-induced $V_{TH}$ variation, as well as the $V_{TH}$ lowering is revealed.

• 한국안전학회지
• /
• 제22권6호
• /
• pp.46-54
• /
• 2007

The purpose of this study was to examines the burning rate of fire retardant treated wood in the cone heater with a one-dimensional integral model. The wood samples used in this study were four species. The species of woods are Redwood, White oak, Douglas fir and Maple. Each sample was nominally 50mm thick and 100mm square. Samples were exposed to a range of incident heat fluxes 10 to $35kW/m^2$ using the cone heater. A one-dimension integral model has been used to predict burning rate, heat of gasification, flame heat fluxes, charring rate and char depth of samples. As a result measurement of mass loss rate, softwoods(Redwood and Douglas fir) has relatively low value than those for hardwoods(White oak and Maple). Average charring rate of woods in case of fire retardant treatment showed reduction effect of 41.29%, 50.00%, 48.18% and 60.82% for Redwood, Douglas fir, White fir and Maple, respectively. Almost all the predictions from integral model showed faster charring than those measured. Average difference between predictions and experimental data was 16%, 9.5% and 11.8% for N, F1 and F2 respectively. Water-soluble fire retardant used in this study find out more effect in hardwood than softwood from the result of measurement of mass loss rate and average charring rate.

• 대한원격탐사학회지
• /
• 제6권2호
• /
• pp.115-133
• /
• 1990

Study was made on the spectral characteristics of rock samples including bentonites collected from the northern Ulsan area. The geology of the area consists mainly of sediments of the Kyongsang Series and Bulguksa granite, the Tertiary volcanics, andesites and tuffs. Relative reflectances of meshed samples(2.5~10mm) to BaSO$_4$ are measured at 6 Landsat TM spectral windows (excluding the thermal band) with HHRR, and their reflection charactristics were analysed. In addition, three different data selection schemes including the Eulidean distance, multiple regression, and PCA weight methods were applied to the 30 TM ratio channels, derived from the above 6 bands. The selected data sets were subject to two unsupervised classification techniques(FA and ISODATA) in order to compare the effectiveness for classification of particularly bentonite from others. As a result, in ISODATA analysis the multiple regression model shows the best, followed by the Euliean distances one. The PCA weight model seems to show some confusion. In FA, though difficult for quantitative analysis, the best still seems to be the regression model. Among ratio bands, rations of band 7 or 5 against other bands represent the best contribution in classification of bentonites from others.

• 정보보호학회논문지
• /
• 제29권1호
• /
• pp.117-125
• /
• 2019

최근 다양한 분야에서 활용중인 딥러닝은 적대적 공격 가능성의 발견으로 위험성이 제기되고 있다. 본 논문에서는 딥러닝의 이미지 분류 모델에서 악의적 공격자가 생성한 적대적 샘플에 의해 분류 정확도가 낮아짐을 실험적으로 검증하였다. 대표적인 이미지 샘플인 MNIST데이터 셋을 사용하였으며, 텐서플로우와 파이토치라이브러리를 사용하여 만든 오토인코더 분류 모델과 CNN(Convolution neural network)분류 모델에 적대적 샘플을 주입하여 탐지 정확도를 측정한다. 적대적 샘플은 MNIST테스트 데이터 셋을 JSMA(Jacobian-based Saliency Map Attack)방법으로 생성한 방법과 FGSM(Fast Gradient Sign Method)방식으로 변형하여 생성하였으며, 분류 모델에 주입하여 측정하였을 때 최소 21.82%에서 최대 39.08%만큼 탐지 정확도가 낮아짐을 검증하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.771-789
• /
• 2019

Text detection has been a popular research topic in the field of computer vision. It is difficult for prevalent text detection algorithms to avoid the dependence on datasets. To overcome this problem, we proposed a novel unsupervised text detection algorithm inspired by bootstrap learning. Firstly, the text candidate in a novel form of superpixel is proposed to improve the text recall rate by image segmentation. Secondly, we propose a unique text sample selection model (TSSM) to extract text samples from the current image and eliminate database dependency. Specifically, to improve the precision of samples, we combine maximally stable extremal regions (MSERs) and the saliency map to generate sample reference maps with a double threshold scheme. Finally, a multiple kernel boosting method is developed to generate a strong text classifier by combining multiple single kernel SVMs based on the samples selected from TSSM. Experimental results on standard datasets demonstrate that our text detection method is robust to complex backgrounds and multilingual text and shows stable performance on different standard datasets.

• Geosystem Engineering
• /
• 제21권5호
• /
• pp.262-272
• /
• 2018

The U-statistic method is one of the most important structural methods to separate the anomaly from background. It considers the location of samples and carries out the statistical analysis of the data without judging from a geochemical point of view and tries to separate subpopulations and determine anomalous areas. In the present study, 3D U-statistic method has been applied for the first time through the three-dimensional (3D) modeling of an ore deposit. In order to achieve this purpose, 3D U-statistic is applied on the data (Fe grade) resulted from the drilling network in Baghak mine, central part of the Sangan iron mines (in Khorassan Razavi Province, Iran). Afterward, results from applying 3D U-statistic method are used for 3D modeling of the iron mineralization. Results show that the anomalous values are well separated from background so that the determined samples as anomalous are not dispersed and according to their positioning, denser areas of anomalous samples could be considered as anomaly areas. And also, final results (3D model of iron mineralization) show that output model using this method is compatible with designed model for mining operation. Moreover, seen that U-statistic method in addition for separating anomaly from background, could be very efficient for the 3D modeling of different ore type.