• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.189-192
• /
• 2009

This paper presents comparison and analysis of various CFA patterns which are used in single sensor digital camera. There are several patterns which are already used, however, images are sometimes darker or brighter than what human see in real life. Also, images show some noise and blurring. To overcome this problems, many studies on the patterns have been discussed. We carry out experiment with seven patterns including the Bayer pattern. The bilinear method is selected for a interpolation method. The experimental result indicates that image quality is not affected by individual patterns and each pattern requires its own interpolation method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2078-2093
• /
• 2019

Visual smoke recognition is a challenging task due to large variations in shape, texture and color of smoke. To improve performance, we propose a novel smoke recognition method by combining dual-encoded features that are extracted from both spatial and Curvelet domains. A Curvelet transform is used to filter an image to generate fifty sub-images of Curvelet coefficients. Then we extract Local Binary Pattern (LBP) maps from these coefficient maps and aggregate histograms of these LBP maps to produce a histogram map. Afterwards, we encode the histogram map again to generate Dual-encoded Local Binary Patterns (Dual-LBP). Histograms of Dual-LBPs from Curvelet domain and Completed Local Binary Patterns (CLBP) from spatial domain are concatenated to form the feature for smoke recognition. Finally, we adopt Gaussian Kernel Optimization (GKO) algorithm to search the optimal kernel parameters of Support Vector Machine (SVM) for further improvement of classification accuracy. Experimental results demonstrate that our method can extract effective and reasonable features of smoke images, and achieve good classification accuracy.

• Journal of Multimedia Information System
• /
• v.6 no.2
• /
• pp.49-60
• /
• 2019

Plants are very crucial for life on Earth. There is a wide variety of plant species available, and the number is increasing every year. Species knowledge is a necessity of various groups of society like foresters, farmers, environmentalists, educators for different work areas. This makes species identification an interdisciplinary interest. This, however, requires expert knowledge and becomes a tedious and challenging task for the non-experts who have very little or no knowledge of the typical botanical terms. However, the advancements in the fields of machine learning and computer vision can help make this task comparatively easier. There is still not a system so developed that can identify all the plant species, but some efforts have been made. In this study, we also have made such an attempt. Plant identification usually involves four steps, i.e. image acquisition, pre-processing, feature extraction, and classification. In this study, images from Swedish leaf dataset have been used, which contains 1,125 images of 15 different species. This is followed by pre-processing using Gaussian filtering mechanism and then texture and color features have been extracted. Finally, classification has been done using Multiclass-support vector machine, which achieved accuracy of nearly 93.26%, which we aim to enhance further.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.23 no.2
• /
• pp.43-52
• /
• 2021

This case study developed and operated "Fashion & Film" as a fashion-related liberal arts course. The class was designed to include fashion styles exhibitted in films, fashion-related PPL, fashion design through film, fashion images of movie characters, considering gender & color images expressed in movie costumes, and the history of western costumes and asian folk costumes in films. The class was counducted through various teaching methods, such as lectures, student's presentation of movie plots, and team discussions, which created a student-led class. The team presentations at the end of the term were intended to enhance the understanding of fashion through movies. The results of subjective lecture evaluation of "Fashion & Film" showed the most satisfaction with the communication with professor. Students said that it was good to understand fashion through film. They expressed a burden with the team project; however, they were satisfied with the team project outcomes. Students said that PowerPoint was used very effectively. On the other hand, there was an prevelent opinion that the content of PowerPoint and workbook did not match. To address this inconvenience, a textbook called "Fashion in Film" was published and used in the first semester of 2020. The multiple-choice evaluation showed that students were generally satisfied with the "Fashion & Film" class.

• Journal of IKEEE
• /
• v.25 no.1
• /
• pp.139-147
• /
• 2021

This paper introduces an image decomposition-based deep learning method and loss function to improve low-light images. In order to remove color distortion and halo artifact, illuminance channel of an input image is decomposed into reflectance and luminance channels, and a decomposition-based multiple structural deep learning process is applied to each channel. In addition, a mixed norm-based loss function is described to increase the stability and remove blurring in reconstructed image. Experimental results show that the proposed method effectively improve various low-light images.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.36.1-37
• /
• 2021

Six images of IRC+10216 taken by the Hubble Space Telescope at three epochs in 2001, 2011, and 2016 are compared in the rest frame of the central carbon star. An accurate astrometry has been achieved with the help of Gaia Data Release 2. The positions of the carbon star in the individual epochs are determined using its known proper motion, defining the rest frame of the star. In 2016, a local brightness peak with compact and red nature is detected at the stellar position. A comparison of the color maps between 2016 and 2011 epochs reveals that the reddest spot moved along with the star, suggesting a possibility of its being the dusty material surrounding the carbon star. Relatively red, ambient region is distributed in an Ω shape and well corresponds to the dusty disk previously suggested based on near-infrared polarization observations. In a larger scale, differential proper motion of multiple ring-like pattern in the rest frame of the star is used to derive the average expansion velocity of transverse wind components, resulting in ~12.5 km s-1 (d/123 pc), where d is the distance to IRC+10216. Three dimensional geometry is implied from its comparison with the line-of-sight wind velocity determined from half-widths of submillimeter emission line profiles of abundant molecules. Uneven temporal variations in brightness for different searchlight beams and anisotropic distribution of extended halo are revisited in the context of the stellar light illumination through a porous envelope with postulated longer-term variations for a period of 10 years.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.38.3-38.3
• /
• 2019

We present newly discovered dwarf galaxy candidates in deep wide-field images of NGC 1291 obtained with KMTNet. We identify 15 dwarf galaxy candidates by visual inspection within the virial radius of NGC 1291. Using imaging simulations, we demonstrate that our imaging data is complete up to 26 mag arcsec^-2 or -10 abs.mag with > 70% of the completeness rate. We also apply automated detection method to find the dwarfs. However, the completeness and the reliability are relatively low compared to the visual inspection. We find that structural and photometric properties of dwarf candidates such as effective radius, central surface brightness, Sérsic index, and absolute magnitude appear to be consistent with those of known dwarf galaxies in nearby groups and clusters, except for color. NGC 1291, residing in a relatively isolated environment, tends to accompany bluer dwarf galaxies (≃0.58) than those in denser environment. It shows that the quenching of dwarfs is susceptible to the environment.

• The Bulletin of The Korean Astronomical Society
• /
• v.45 no.1
• /
• pp.66.4-67
• /
• 2020

Gyeonggi science high school for the gifted (GSHS) installed 60cm telescope, which is waiting for student observers. It is essential to understand the characteristics of the photometric system, consisting of telescope, filter, and CCD, to get reliable data. CCD images of SA98 Landolt standard field and M67 were obtained on 19th March 2020. The images of each field were combined by filters, i.e., we ignored the monochromatic atmospheric extinction since the photometric objects themselves are standard stars. 24 standard stars in SA98 field and 12 standard stars in M67 were used to derive the tentative transformation equation between our bv photometric system and Johnson BV photometric system. In this poster, we present the preliminary standardization result for Johnson BV photometric system in GSHS 60cm telescope. The reproductivity is discussed by comparing color coefficients of two fields. We plan to extend this process to Johnsons-Cousins BVRI photometric system and narrow-band filters for flux calibration.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.31.2-31.2
• /
• 2021

To estimate the contribution of quasars on keeping the IGM ionized, building a quasar luminosity function (LF) is necessary. Quasar LFs derived from multiple quasar surveys, however, are incompatible, especially for the faint regime, emphasizing the need for deep images. In this study, we construct quasar LF reaching M₁₄₅₀~-21.5 AB magnitude at z ~ 5, which is 1.5 mag deeper than previously reported LFs, using deep images from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). We trained an artificial neural network (ANN) by inserting the colors as inputs to classify the quasars at z ~ 5 from the late-type stars and low-redshift galaxies. The accuracy of ANN is > 99 %. We also adopted the Bayesian information criterion to elaborate on the quasar-like objects. As a result, we recovered 5/5 confirmed quasars and remarkably minimized the contamination rate of high-redshift galaxies by up to six times compared to the selection using color selection alone. The constructed quasar parametric LF shows a flatter faint-end slope α=-127^+0.16_-0.15 similar to the recent LFs. The number of faint quasars (M₁₄₅₀ < -23.5) is too few to be the main contributor to IGM ionizing photons.

• Journal of Platform Technology
• /
• v.11 no.6
• /
• pp.13-20
• /
• 2023

Regularly inspecting vehicle tires' condition is imperative for driving safety and comfort. Poorly maintained tires can pose fatal risks, leading to accidents. Unfortunately, manual tire visual inspections are often considered no less laborious than employing an automatic tire inspection system. Nevertheless, an automated tire inspection method can significantly enhance driver compliance and awareness, encouraging routine checks. Therefore, there is an urgency for automated tire inspection solutions. Here, we focus on developing a deep learning (DL) model to predict cracked tires. The main idea of this study is to demonstrate the comparative analysis of DenseNet121, VGG-19 and EfficientNet Convolution Neural Network-based (CNN) Transfer Learning (TL) and suggest which model is more recommended for cracked tire classification tasks. To measure the model's effectiveness, we experimented using a publicly accessible dataset of 1028 images categorized into two classes. Our experimental results obtain good performance in terms of accuracy, with 0.9515. This shows that the model is reliable even though it works on a dataset of tire images which are characterized by homogeneous color intensity.