• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.81.2-81.2
• /
• 2019

We develop an image-to-image translation model, which is a popular deep learning method based on conditional Generative Adversarial Networks (cGANs), to generate solar magnetograms and EUV images from sunspot drawings. For this, we train the model using pairs of sunspot drawings from Mount Wilson Observatory (MWO) and their corresponding SDO/HMI magnetograms and SDO/AIA EUV images (512 by 512) from January 2012 to September 2014. We test the model by comparing pairs of actual SDO images (magnetogram and EUV images) and the corresponding AI-generated ones from October to December in 2014. Our results show that bipolar structures and coronal loop structures of AI-generated images are consistent with those of the original ones. We find that their unsigned magnetic fluxes well correlate with those of the original ones with a good correlation coefficient of 0.86. We also obtain pixel-to-pixel correlations EUV images and AI-generated ones. The average correlations of 92 test samples for several SDO lines are very good: 0.88 for AIA 211, 0.87 for AIA 1600 and 0.93 for AIA 1700. These facts imply that AI-generated EUV images quite similar to AIA ones. Applying this model to the Galileo sunspot drawings in 1612, we generate HMI-like magnetograms and AIA-like EUV images of the sunspots. This application will be used to generate solar images using historical sunspot drawings.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.6_2
• /
• pp.1025-1032
• /
• 2020

In this paper, we propose a method to classify road weather conditions into rain, fog, and sun using a SVM (Support Vector Machine) classifier after extracting weather features from images acquired in real time using an optical sensor installed on a roadside post. A multi-dimensional weather feature vector consisting of factors such as image sharpeness, image entropy, Michelson contrast, MSCN (Mean Subtraction and Contrast Normalization), dark channel prior, image colorfulness, and local binary pattern as global features of weather-related images was extracted from road images, and then a road weather classifier was created by performing machine learning on 700 sun images, 2,000 rain images, and 1,000 fog images. Finally, the classification performance was tested for 140 sun images, 510 rain images, and 240 fog images. Overall classification performance is assessed to be applicable in real road services and can be enhanced further with optimization along with year-round data collection and training.

• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.21 no.2
• /
• pp.133-138
• /
• 2010

Background and Objectives : High-speed imaging can be useful in studies of linguistic and artistic singing styles, and laryngeal examination of patients with voice disorders, particularly in irregular vocal fold vibrations. In this study, we introduce new laryngeal imaging systems which are commercially available high speed cameras connected with a laryngoscope. Materials and Method : The laryngeal images were captured from three different types of cameras. First, the adapter was made to connect with laryngoscope and Casio EX-F1 to capture the images using $2{\times}150$ Watt Halogen light source (EndoSTROB) at speeds of 1,200 tps (frame per second)($336{\times}96$). Second, Phantom Miro ex4 was used to capture the digital laryngeal images using Xenon Nova light source 175 Watt (STORZ) at speeds of 1,920 fps ($512{\times}384$). Finally, laryngeal images were captured using MotionXtra N-4 with 250 Watt halogen lamp (Olympus CLH-250) light source at speeds of 2,000tps ($384{\times}400$) by connecting with laryngoscope. All images were transformed into the Kymograph using KIPS (Kay's image processing Software) of Kay Pentex Inc. Results: Casio EX-F1 was too small to adjust the focus and screen size was diminished once the images were captured despite of high resolution images. High quality of color images could be obtained with Phantom Miro ex4 whereas good black and white images from Motion Xtra N-4 Despite of some limitations of illumination problems, limited recording time capacity, and time consuming procedures in Phantom Miro ex4 and Motion Xtra N-4, those portable devices provided high resolution images. Conclusion : All those high speed cameras could capture the laryngeal images by connecting with laryngoscope. High resolution images were able to be captured at the fixed position under the good lightness. Accordingly, these techniques could be applicable to observe the vocal fold vibration properties in the clinical practice.

• Journal of the Semiconductor & Display Technology
• /
• v.14 no.4
• /
• pp.13-19
• /
• 2015

In this paper, the principle of devices which can magnify images without distorting the images is described. When the device is put on a smart phones, viewers can see the magnified images. Magnified images can be few 100 times bigger than the original images. Therefore, viewers can see movie theater size images with the device put on a smart phone. Two different schemes are explained in the paper and one realization of the device is presented. The device can be used in many different application areas.

• Science of Emotion and Sensibility
• /
• v.14 no.3
• /
• pp.403-414
• /
• 2011

The current study examined brain regions associated with aesthetic experience to fractal images using functional MRI. The aesthetic estimations of the images showed that there is a general consensus regarding the perception of beautiful images. Out of 270 fractal images, fifty images rated highest(beautiful images) and fifty images rated lowest(non-beautiful images) were selected and presented to the participants. The two conditions were presented using the block design. Frontal lobes, cingulate gyri, and insula, the areas related to the cognitive and emotional processing in aesthetic experience, were activated when beautiful images were presented. In contrast, the middle occipital gyri and precuneus, the areas associated with experience of negative emotions, were activated when non-beautiful images were presented. The conjunction analysis showed activations in temporal areas in response to beautiful images and activations in parietal areas in response to non-beautiful images. These results indicate that beautiful images elicit semantic interpretations whereas non-beautiful images facilitate abstract processes.

• Journal of Biomedical Engineering Research
• /
• v.42 no.4
• /
• pp.167-174
• /
• 2021

The purpose of this study is to classify TIFF images, PNG images, and JPEG images using deep learning, and to compare the accuracy by verifying the classification performance. The TIFF, PNG, and JPEG images converted from chest X-ray DICOM images were applied to five deep neural network models performed in image recognition and classification to compare classification performance. The data consisted of a total of 4,000 X-ray images, which were converted from DICOM images into 16-bit TIFF images and 8-bit PNG and JPEG images. The learning models are CNN models - VGG16, ResNet50, InceptionV3, DenseNet121, and EfficientNetB0. The accuracy of the five convolutional neural network models of TIFF images is 99.86%, 99.86%, 99.99%, 100%, and 99.89%. The accuracy of PNG images is 99.88%, 100%, 99.97%, 99.87%, and 100%. The accuracy of JPEG images is 100%, 100%, 99.96%, 99.89%, and 100%. Validation of classification performance using test data showed 100% in accuracy, precision, recall and F1 score. Our classification results show that when DICOM images are converted to TIFF, PNG, and JPEG images and learned through preprocessing, the learning works well in all formats. In medical imaging research using deep learning, the classification performance is not affected by converting DICOM images into any format.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.1 no.1
• /
• pp.53-71
• /
• 1999

The purpose of this study is to determine the fashion images implied in men's fashion trends and systematize the characteristics of the textile materials used for fashion images, by analyzing men's fashion trends published by Korean fashion information service companies. This study would be meaningful if it can suggest some objective criteria for the characteristics of textile per fashion image. The researcher analyzed the data on the basis of 8 fashion images, which were ethnic, modern, traditional, avant-garde, active, romantic, natural, techno ones. Above men's fashion images were choosed by analyze the some literatures and men's fashion trend information. The data used for this study were information about S/S and F/W men's fashion trends published by Interfashion planning, Samsung fashion Research Center for the period of 1995-2000. The data collected were subject to “content analysis method”. As a result of the analysis, the major images of 1995-2000 were natural, active, traditional, modern, ethnic, avant-garde, techno images, and while such combinations of conflicting images as ethnic/modern, traditional/avant-garde, natural/techno. Other mixed images were ethnic/natural, modern/active, tradional/active, traditional/modern, romantic/modern, ethnic/romantic, traditional/natural, modern/natural, active/natural, active/traditional/natural, etc. The various characteristics of eight men's fashion images were found in color, pattern and textile materials.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.11
• /
• pp.4153-4169
• /
• 2014

To recognize F5-like (such as F5 and nsF5) steganographic algorithm from multi-class stego images, a recognition algorithm based on the identifiable statistical feature (IDSF) of F5-like steganography is proposed in this paper. First, this paper analyzes the special modification ways of F5-like steganography to image data, as well as the special changes of statistical properties of image data caused by the modifications. And then, by constructing appropriate feature extraction sources, the IDSF of F5-like steganography distinguished from others is extracted. Lastly, based on the extracted IDSFs and combined with the training of SVM (Support Vector Machine) classifier, a recognition algorithm is presented to recognize F5-like stego images from images set consisting of a large number of multi-class stego images. A series of experimental results based on the detection of five types of typical JPEG steganography (namely F5, nsF5, JSteg, Steghide and Outguess) indicate that, the proposed algorithm can distinguish F5-like stego images reliably from multi-class stego images generated by the steganography mentioned above. Furthermore, even if the types of some detected stego images are unknown, the proposed algorithm can still recognize F5-like stego images correctly with high accuracy.

• Imaging Science in Dentistry
• /
• v.48 no.1
• /
• pp.41-44
• /
• 2018

Purpose: To evaluate the effect of different numbers of basis images and the use of metal artifact reduction (MAR) on the production and reduction of artifacts in cone-beam computed tomography images. Materials and Methods: An acrylic resin phantom with a metal alloy sample was scanned, with 450 or 720 basis images and with or without MAR. Standard deviation values for the test areas (around the metal object) were obtained as a way of measuring artifact production. Two-way analysis of variance was used with a 5% significance level. Results: There was no significant difference in artifact production among the images obtained with different numbers of basis images without MAR (P=.985). MAR significantly reduced artifact production in the test areas only in the protocol using 720 basis images (P=.017). The protocol using 450 basis images with MAR showed no significant difference in artifact production when compared to the protocol using 720 basis images with MAR (P=.579). Conclusion: Protocols with a smaller number of basis images and with MAR activated are preferable for minimizing artifact production in tomographic images without exposing the patient to a greater radiation dose.

• Korean Institute of Interior Design Journal
• /
• no.21
• /
• pp.17-25
• /
• 1999

This study attempts to analyze the styles and the images of the exterior and interior design of modern Korean houses since 1960s, and to understand the characteristics of Korean design by analyzing the consistant and changing factors in the styles and images of modern Korean houses. The photos of 101 houses were used for content-analysis approach. The result shows that the most dominant exterior style of modern Korean houses has been changed from modernism to late-modernism and it brought some changes in the images with increase of deconstructivism and post-modernism style. While modernism has been consistantly the most dominant style in the interior design of modern Korean houses with unified, urbane, masculine and unpartitioned images. It means that modernism style shown in modern Korean houses expresses certain consistant images, which can be defined as a characteristic of modern Korean house desigv.