• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.148-153
• /
• 2009

Few methods have dealt with segmenting multiple images with analogous content. Concurrent images of a scene and gathered images of a similar foreground are examples of these images, which we term consistent scene images. In this paper, we present a method to segment these images based on manual segmentation of one image, by iteratively propagating information via multi-level cues with adaptive confidence. The cues are classified as low-, mid-, and high- levels based on whether they pertain to pixels, patches, and shapes. Propagated cues are used to compute potentials in an MRF framework, and segmentation is done by energy minimization. Through this process, the proposed method attempts to maximize the amount of extracted information and maximize the consistency of segmentation. We demonstrate the effectiveness of the proposed method on several sets of consistent scene images and provide a comparison with results based only on mid-level cues [1].

• Journal of Korean Elementary Science Education
• /
• v.28 no.1
• /
• pp.35-45
• /
• 2009

We analyzed $3^{rd}$ graders' images of scientists and science learning students. We chose $3^{rd}$ graders because this is the time when children first encounter formal science learning opportunities. Draw-A-Scientist-Test (DAST) and the revised Draw-A-Scientist-Test Checklist (DAST-C) were used to analyze students' images of scientists, whereas Drawing-A-Science-Learner- and a checklist were used to analyze students' images of science learning students. We found that $3^{rd}$ graders showed common features of scientists who wore laboratory coats but not wearing glasses, goggles or masks and smiling. While most boys drew a male scientist, about a half of girls drew female scientists. Old and weird looking images of scientists that were typically known in other literatures were not found in this study. Science learning students were not wearing lab coasts, glasses, goggles, nor masks. Most of those students were conducting chemistry related experiments, which seemed to be influenced by the $3^{rd}$ grade's science curriculum. We also found relationships among components of images of scientists and science learning students. Although $3^{rd}$ graders' images of scientists and science learning students showed common features, this typical image was not the same as the previous studies have reported. This implies that the images of scientists and science learning students have not yet fixed by $3^{rd}$ grade. Thus, this seems to be a critical time when children start developing images of scientists. Children's direct experiences in the science classroom along with environmental factors such as media exposures can influence their formation of images of scientists and science learning students.

• The Korean Journal of Nuclear Medicine
• /
• v.29 no.4
• /
• pp.478-483
• /
• 1995

We compared the influences of reconstruction methods using $180^{\circ}$ or $360^{\circ}$ data upon contrasts and discriminating capability and diagnostic accuracy in $^{99m}Tc-MIBI$ stress/rest myocardial SPECT. We reviewed SPECT images reconstructed only with $180^{\circ}$ projection data or with $360^{\circ}$ data in 18 patients and in 11 normal subjects. To compare counts of surface structures and deep structures, we measured ape# posterior wall ratios in 11 normal subjects. To compare the contrasts of images, we measured apex/ventricle ratios. To compare contrasts between normal and diseased myocardial segments, we measured count ratios of defect and normal segments in 4 patients who had single coronary artery diseases. To compare diagnostic accuracy, we scored SPECT images made with $180^{\circ}$ and $360^{\circ}$ data segmentally. Sensitivity and specificity for the diagnosis of coronary artery disease and for the revelation of diseased arteries with both $180^{\circ}$ and $360^{\circ}$ SPECT images. If involved coronary arteries had more narrowing than 50% In coronary angiogram, we considered them as diseased arteries Apex/posterior wall ratios were not different significantly in normal subjects. Apex/ ventricle ratios in normal subjects were different significantly between $180^{\circ}$ and $360^{\circ}$ SPECT images. Defect/normal ratios were different significantly between $180^{\circ}$ and $360^{\circ}$ SPECT images in single vessel disease patients. The overall diagnostic accurracy was the same between $180^{\circ}$ and $360^{\circ}$ data collection. Sensitivity was 94% and specificity was 91% for both types of data collection in this sample population. Sensitivity and specificity of each coronary artery territory were not significantly different between the images made with $180^{\circ}$ and $360^{\circ}$ data. The images made with $180^{\circ}$ data had better contrast between ventricle and myocardium and between hypoperfused and normal myocardium, though no difference was found between the ratios of the myocardial counts of surface and deep structures. However, diagnostic sensitivities of diseased artery territories were not different significantly and so were overall diagnostic accuracy between both methods of making images with $180^{\circ}$ and $360^{\circ}$ data.

• Journal of Biomedical Engineering Research
• /
• v.15 no.1
• /
• pp.57-62
• /
• 1994

The demand of image compression is increasing now for the integration of medical images into the hospital information system. Even though the quantitative distortion can be measured from the difference between original and reconstructed images, it doesn't include the nonlinear characteristics of human visual system. In this study, we have evaluated the nonlinear characteristics of human visual system and applied them to the compression of medical images. The distortion measures which reflect the characteristics of human visual system has been considered. This image compression procedure consists of coding scheme using JND (Just Noticeable Difference) curve, polynomial approximation and BTC (Block Truncation Coding). Results show that this method can be applied to CT images, scanned film images and other kinds of medical images with the compression ratio of 5-10:1 without any noticeable distortion.

• Journal of Biomedical Engineering Research
• /
• v.8 no.2
• /
• pp.171-176
• /
• 1987

Wiener filtering method was applied to the dual energy imaging procedure in digital radiography(D.R.). A linear scanning photodiode arrays with 1024 elements(0.6mm H 1.3mm pixel size) were used to obtain chest images in 0.7 sec. For high energy image acquisition, X-ray tube was set at 140KVp, 100mA with a rare-earth phosphor screen. Low energy image was obtained with X-ray tube setting at 70KVp, 150mA. These measured dual energy images are represented in the vector matrix notation as a linear discrete model including the additive random noise. Then, the object images are restored in the minimum mean square error sense using Wiener filtering method in the transformed domain. These restored high and low energy images are used for computation of the basis image decomposition. Then the basis images are linearly combined to produce bone or tissue selective images. Using this process, we could improve the signal to noise ratio characteristics in the material selective images.

• Journal of the Optical Society of Korea
• /
• v.18 no.3
• /
• pp.217-224
• /
• 2014

A major advantage of wavefront reconstruction based on a series of diffracted intensity images using only single-beam illumination is the simplicity of setup. Here we propose a fast-converging algorithm for wavefront calculation using single-beam illumination. The captured intensity images are resampled to a series of intensity images, ranging from highest to lowest resampling; each resampled image has half the number of pixels as the previous one. Phase calculation at a lower resolution is used as the initial solution phase at a higher resolution. This corresponds to separately calculating the phase for the lower- and higher-frequency components. Iterations on the low-frequency components do not need to be performed on the higher-frequency components, thus making the convergence of the phase retrieval faster than with the conventional method. The principle is verified by both simulation and optical experiments.

• Journal of Korea Multimedia Society
• /
• v.22 no.12
• /
• pp.1447-1456
• /
• 2019

As artificial intelligence technology is actively used in image processing, it is possible to generate high-quality fake images based on deep learning. Fake images generated using GAN(Generative Adversarial Network), one of unsupervised learning algorithms, have reached levels that are hard to discriminate from the naked eye. Detecting these fake images is required as they can be abused for crimes such as illegal content production, identity fraud and defamation. In this paper, we develop a deep-learning model based on CNN(Convolutional Neural Network) for the detection of StyleGAN fake images. StyleGAN is one of GAN algorithms and has an excellent performance in generating face images. We experiment with 48 number of experimental scenarios developed by combining parameters of the proposed model. We train and test each scenario with 300,000 number of real and fake face images in order to present a model parameter that improves performance in the detection of fake faces.

• Journal of Korea Multimedia Society
• /
• v.20 no.7
• /
• pp.994-1003
• /
• 2017

Ground-glass opacity nodules(GGNs) in chest CT images are associated with lung cancer, and have a different malignant rate depending on existence of solid component in the nodules. In this paper, we propose a method to classify pure GGNs and part-solid GGNs using multiview images and texture analysis in pulmonary GGNs with solid components of 5mm or smaller. We extracted 1521 features from the GGNs segmented from the chest CT images and classified the GGNs using a SVM classification model with selected features that classify pure GGNs and part-solid GGNs through a feature selection method. Our method showed 85% accuracy using the SVM classifier with the top 10 features selected in the multiview images.

• The Korean Journal of Nuclear Medicine
• /
• v.23 no.1
• /
• pp.19-25
• /
• 1989

To assess the usefulness of myocardial SPECT imaging to detect post-myocardial infarction ventricular aneurysms, we analyzed the Technetium-99m MIBI myocardial SPECT images of 16 patients with anterior and/or apical infarction, 9 had the previously reported findings of failure of convergence of the left ventricular walls toward the apex on SPECT images and 8 of them also had ventricular aneurysms. The ventriculography of the 2 patients with mixed pattern revealed 1 case of ventricular aneurysm and 1 case without aneurysm. Among the remaining 5 pateints with converging pattern, none had ventricular aneurysm. Of the other 11 pateints with inferior and/or lateral wall infarction, 1 patient had ventricular aneurysm and the SPECT image couldn't detect the aneurysm. $Department of Internal Medicine, College of Medicine, Seoul National University$ myocardial SPECT images for the detection of ventricular aneurysm had a sensitivity of 90 %, a specificity of 88%, and an accuracy of 89%. Thus we could get the information about presence of ventricular aneurysm as well as the status of the myocardial perfusion from the Tc-99m MIBI myocardial SPECT images.

• Imaging Science in Dentistry
• /
• v.42 no.1
• /
• pp.25-33
• /
• 2012

Purpose : This study was performed to determine the accuracy of linear measurements on three-dimensional (3D) images using multi-detector computed tomography (MDCT) and cone-beam computed tomography (CBCT). Materials and Methods : MDCT and CBCT were performed using 24 dry skulls. Twenty-one measurements were taken on the dry skulls using digital caliper. Both types of CT data were imported into OnDemand software and identification of landmarks on the 3D surface rendering images and calculation of linear measurements were performed. Reproducibility of the measurements was assessed using repeated measures ANOVA and ICC, and the measurements were statistically compared using a Student t-test. Results : All assessments under the direct measurement and image-based measurements on the 3D CT surface rendering images using MDCT and CBCT showed no statistically difference under the ICC examination. The measurements showed no differences between the direct measurements of dry skull and the image-based measurements on the 3D CT surface rendering images (P>.05). Conclusion : Three-dimensional reconstructed surface rendering images using MDCT and CBCT would be appropriate for 3D measurements.