• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.6
• /
• pp.1946-1963
• /
• 2014

Most background subtraction methods focus on dynamic and complex scenes without considering robustness against noise. This paper proposes a background subtraction algorithm based on dictionary learning and sparse coding for handling low light conditions. The proposed method formulates background modeling as the linear and sparse combination of atoms in the dictionary. The background subtraction is considered as the difference between sparse representations of the current frame and the background model. Assuming that the projection of the noise over the dictionary is irregular and random guarantees the adaptability of the approach in large noisy scenes. Experimental results divided in simulated large noise and realistic low light conditions show the promising robustness of the proposed approach compared with other competing methods.

• Corrosion Science and Technology
• /
• v.3 no.4
• /
• pp.166-171
• /
• 2004

Recent developments in Bridge Management Systems (BMS) and in Life-Cycle Cost (LCC) of bridges, have raised the need for evaluation procedure of future condition (Deterioration) of a bridge. Predicting future deterioration is not an easy task due to limited past data to extrapolate from and also due to difficulty in measuring actual deterioration such as section loss of steel on an actual steel bridge. Also, increase in live load and reduction of resistance are random variables, thus a probabilistic approach should be adopted for determining the future deterioration. Due to difficulties in evaluation of future deterioration on steel bridges, accepting uncertainties within a reasonable error, a deterministic procedure using bridge condition rating can be a useful tool for projection of future condition of bridges to identify repair and maintenance needs. The object of this paper is to determine applicability of evaluating deterioration of steel bridge components based on Bridge condition ratings. Bridge condition ratings of bridge components show wide variation for bridges of same age and does not directly correlate well with the age of the bridge and/or deterioration of the bridge. High uncertainty can be reduced by breaking down the rating and by sensitivity analysis. From refined condition rating data, generalized deterioration profile of structures based on age can be derived. Examples are shown for sample bridges in USA. Approximately, 3,000 short to medium span steel bridges were listed in the inventory database. Results show wide variation of rating factors but by subdividing the Bridge condition ratings for various categories general deterioration profiles of steel bridges can be determined.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37A no.9
• /
• pp.780-785
• /
• 2012

In the current study, the authors propose a method for extracting license plate regions by means of a neural network trained to output the plates center of gravity. The method is shown to be effective. Since the learning pattern presentation positions are defined by random numbers, a different pattern is submitted to the neural network for learning each time, which enables it to form a neural network with high universality of coverage. The article discusses issues of the optimal learning surface for a license plate covered by the learning pattern, the effect of suppression learning of the number and headlight sections, as well as the effect of learning pattern enlargement/reduction and of concentration value conversion. Results of evaluation tests based on pictures of 595 vehicles taken at an underground parking garage demonstrated detection rates of 98.5%.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.10
• /
• pp.48-58
• /
• 1998

We present a method that can recognize face images using independent component analysis that is used mainly for blind sources separation in signal processing. We assumed that a face image can be expressed as the sum of a set of statistically independent feature images, which was obtained by using independent component analysis. Face recognition was peformed by projecting the input image to the feature image space and then by comparing its projection components with those of stored reference images. We carried out face recognition experiments with a database that consists of various varied face images (total 400 varied facial images collected from 10 per person) and compared the performance of our method with that of the eigenface method based on principal component analysis. The presented method gave better results of recognition rate than the eigenface method did, and showed robustness to the random noise added in the input facial images.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.1 no.3
• /
• pp.105-109
• /
• 2008

In the current study, the authors propose a method for extracting license plate regions by means of a neural network trained to output the plate.s center of gravity. The method is shown to be effective. Since the learning pattern presentation positions are defined by random numbers, a different pattern is submitted to the neural network for learning each time, which enables it to form a neural network with high universality of coverage. The article discusses issues of the optimal learning surface for a license plate covered by the learning pattern, the effect of suppression learning of the number and headlight sections, as well as the effect of learning pattern enlargement/reduction and of concentration value conversion. Results of evaluation tests based on pictures of 595 vehicles taken at an underground parking garage demonstrated detection rates of 98.5%.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.8
• /
• pp.601-605
• /
• 2009

Recently, photo sharing and publishing based Social Network Sites(SNSs) are increasingly attracting the attention of academic and industry researches. Millions of users have integrated these sites into their daily practices to communicate with online people. In this paper, we propose an efficient face annotation and retrieval system under SNS. Since the system needs to deal with a huge database which consists of an increasing users and images, both effectiveness and efficiency are required, In order to deal with this problem, we propose a face annotation classifier which adopts an online learning and social decomposition approach. The proposed method is shown to have comparable accuracy and better efficiency than that of the widely used Support Vector Machine. Consequently, the proposed framework can reduce the user's tedious efforts to annotate face images and provides a fast response to millions of users.

• Journal of Korea Game Society
• /
• v.21 no.2
• /
• pp.43-54
• /
• 2021

The goal of this research is contemplation on whether postmodern design concepts such as parody, simulation, appropriation, pastiche apply to the character design in games, within bounds of postmodernism. In this philosophy, designs display a mixture of advanced culture and mass culture, emphasize philosophical contingency, and appear to be a random collage of ideas. it was perceived that the visual art in game character design is also a projection of the postmodernist philosophy. Upon analyzing this perception, character designs categorized based on the applicability of postmodern designs

• Current Optics and Photonics
• /
• v.5 no.6
• /
• pp.686-698
• /
• 2021

A camouflaged encryption scheme based on Hadamard matrix and ghost imaging is proposed. In the process of the encryption, an orthogonal matrix is used as the projection pattern of ghost imaging to improve the definition of the reconstructed images. The ciphertext of the secret image is constrained to the camouflaged image. The key of the camouflaged image is obtained by the method of sparse decomposition by principal component orthogonal basis and the constrained ciphertext. The information of the secret image is hidden into the information of the camouflaged image which can improve the security of the system. In the decryption process, the authorized user needs to extract the key of the secret image according to the obtained random sequences. The real encrypted information can be obtained. Otherwise, the obtained image is the camouflaged image. In order to verify the feasibility, security and robustness of the encryption system, binary images and gray-scale images are selected for simulation and experiment. The results show that the proposed encryption system simplifies the calculation process, and also improves the definition of the reconstructed images and the security of the encryption system.

• Food Science of Animal Resources
• /
• v.44 no.4
• /
• pp.934-950
• /
• 2024

This study addresses the prevalent issue of meat species authentication and adulteration through a chemometrics-based approach, crucial for upholding public health and ensuring a fair marketplace. Volatile compounds were extracted and analyzed using headspace-solid-phase-microextraction-gas chromatography-mass spectrometry. Adulterated meat samples were effectively identified through principal component analysis (PCA) and partial least square-discriminant analysis (PLS-DA). Through variable importance in projection scores and a Random Forest test, 11 key compounds, including nonanal, octanal, hexadecanal, benzaldehyde, 1-octanol, hexanoic acid, heptanoic acid, octanoic acid, and 2-acetylpyrrole for beef, and hexanal and 1-octen-3-ol for pork, were robustly identified as biomarkers. These compounds exhibited a discernible trend in adulterated samples based on adulteration ratios, evident in a heatmap. Notably, lipid degradation compounds strongly influenced meat discrimination. PCA and PLS-DA yielded significant sample separation, with the first two components capturing 80% and 72.1% of total variance, respectively. This technique could be a reliable method for detecting meat adulteration in cooked meat.

• Journal of Korea Water Resources Association
• /
• v.52 no.5
• /
• pp.337-347
• /
• 2019

This study proposes the framework to select the representative general circulation model (GCM) for climate change projection. The grid-based results of GCMs were transformed to all considered meteorological stations using inverse distance weighted (IDW) method and its results were compared to the observed precipitation. Six quantile mapping methods and random forest method were used to correct the bias between GCM's and the observation data. Thus, the empirical quantile which belongs to non-parameteric transformation method was selected as a best bias correction method by comparing the measures of performance indicators. Then, one of the multi-criteria decision techniques, TOPSIS (Technique for Order of Preference by Ideal Solution), was used to find the representative GCM using the performances of four GCMs after the bias correction using empirical quantile method. As a result, GISS-E2-R was the best and followed by MIROC5, CSIRO-Mk3-6-0, and CCSM4. Because these results are limited several GCMs, different results will be expected if more GCM data considered.