• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.798-801
• /
• 2014

Nowadays, the demand for multimedia services has grown with the rapid evolution in the digital era. But due to external causes in the process of processing, transmitting and storing image data, the images are damaged. One of the major causes of such damage is known to be noise. Some of the most commonly used methods for removing noise are CWMF(center weighted median filter), A-TMF(alpha-trimmed mean filter) and AWMF(adaptive weighted median filter). However, these filters all leave a bit to be desired in removing noise in a complex noise environment. Therefore this paper suggest an image restoration filter algorithm that first judges the noise and sets a adjustment weight based on the median value and distance of the mask to remove the complex noise. For an objective analysis, the results were compared against existing methods and the PSNR(peak signal to noise ratio) was used as a reference.

• English & American cultural studies
• /
• v.10 no.2
• /
• pp.149-207
• /
• 2010

Written simultaneously with Prometheus Unbound, Shelley's "Julian & Maddalo" is a masterwork of dramatic poiesis, of doubling embedded in its couplets, dialogic debate on human nature and contrasted symbolic emblems. The emblems mirror each other and are themselves sites of generative paradox: the "heaven illumined" but "dreary tower" of the Maniac and the glorious sunsets on the "ever-shifting sand" of the Lido, a wasteland that is a place of self discovery but also of "abandonment" and barren mingling figured, inter alia, in its "amphibious weeds," a trope of the poem's personae. This essay also explores the poem's dramatic structure and various rhetorical devices, beginning with the Preface, a threshold of complex identity disguise that Shelley uses for veiled self-presentation, as in "Alastor," mirroring and literary references replete with nuanced ironies. I focus mainly on the complex figures of liminality Shelley uses to develop his own thoughts (as well as his ongoing debates with Byron) about man's potential for growth in thought, insight and empathy, in political reform and interpersonal and individual healing. Advancing Shelley's most optimistic ideas, Julian, escorted by Maddalo observes the Maniac, -- a living ruin whose pained eloquence reveals the link of eros to poiesis and the limits of the latter's ability to 'transform a world.' The Maniac is the core of muse-work (remembering, thinking and song) and Shelley presents him as its emblem. He also is prefigured in and reflects the quintessentially liminal Lido with its "barren embrace" of sea and land. Yet it is less the Maniac's feeling that his grief is "charactered in vain…on this unfeeling leaf" than Julian's rationales for leaving the site of pain that point to Shelley's final comment on poetry's transformative limits. As the primary haploids of the drama's meiosis re-combine and two of them, Maddalo and the maniac fall away, an analogy I briefly develop and embedded in the erotic dynamics of poiesis, Shelley suggests, as he did at the beginning of his poetic lyricism in "Alastor" and at its end in "the Triumph of Life"that images mislead and delude; that "the deep truth is imageless" and redemption is not in but beyond figuration.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.5
• /
• pp.1-8
• /
• 2021

Coronavirus disease 2019 (COVID-19) has affected the world seriously. Every person is required for wearing a mask properly in a public area to prevent spreading the virus. However, many people are not wearing a mask properly. In this paper, we propose an efficient mask detection system. In our proposed system, we first detect the faces of input images using YOLOv5 and classify them as the one of three scene complexity classes (Simple, Moderate, and Complex) based on the number of detected faces. After that, the image is fed into the Faster-RCNN with the one of three ResNet (ResNet-18, 50, and 101) as backbone network depending on the scene complexity for detecting the face area and identifying whether the person is wearing the mask properly or not. We evaluated our proposed system using public mask detection datasets. The results show that our proposed system outperforms other models.

• Imaging Science in Dentistry
• /
• v.52 no.1
• /
• pp.11-18
• /
• 2022

Purpose: This study aimed to assess incidental abnormal findings in the paranasal sinuses and anatomical variations of the ostiomeatal complex (OMC) on magnetic resonance imaging (MRI) scans. Materials and Methods: MRI scans of 616 patients (mean age, 44.0±19.4 years) were evaluated. Prior to obtaining the MRI scans, a checklist of patients' clinical symptoms was filled out after obtaining their consent. The Lund-Mackay classification was used to assess the paranasal sinuses and OMC. The prevalence of abnormal findings and their associations with patients' age, sex, and subjective symptoms were analyzed by the chi-square test, independent-sample t-test, and analysis of variance. The level of significance was set at 0.05. Results: Abnormal findings in the paranasal sinuses were detected in 32.0% of patients, with a significantly higher prevalence in males (P<0.05), but no significant association with age (P>0.05). Epithelial thickening and retention cyst were the most common abnormal findings in the paranasal sinuses. According to the Lund-Mackay classification, 93% of the study population had normal sinuses (score<4). Concha bullosa and paradoxical concha were detected in 15.3% and 3.4%, respectively, with no significant association with the presence of septal deviation or Lund-Mackay classification (P>0.05). Conclusion: Considering the relatively high prevalence of abnormal findings in the paranasal sinuses, it appears that clinical symptoms alone are not sufficient to diagnose sinusitis. A more accurate strategy would be to assess radiographic images of the paranasal sinuses and use a classification system. Sinusitis should be suspected in patients receiving a high score in this classification.

• Journal of Biomedical Engineering Research
• /
• v.44 no.6
• /
• pp.377-383
• /
• 2023

Purpose: There is increasing attention to the application of transcranial direct current stimulation (tDCS) for enhancing cognitive functions in subjects to aging, mild cognitive impairment (MCI), and Alzheimer's disease (AD). Despite varying treatment outcomes in tDCS which depend on the amount of current reaching the brain, there is no general information on the impacts of anatomical features associated with AD on tDCS-induced electric field. Objective: The objective of this study is to examine how AD-related anatomical variation affects the tDCS-induced electric field using computational modeling. Methods: We collected 180 magnetic resonance images (MRI) of AD patients and healthy controls from a publicly available database (Alzheimer's Disease Neuroimaging Initiative; ADNI), and MRIs were divided into female-AD, male-AD, female-normal, and male-normal groups. For each group, segmented brain volumes (cerebrospinal fluid, gray matter, ventricle, rostral middle frontal (RMF), and hippocampus/amygdala complex) using MRI were measured, and tDCS-induced electric fields were simulated, targeting RMF. Results: For segmented brain volumes, significant sex differences were observed in the gray matter and RMF, and considerable disease differences were found in cerebrospinal fluid, ventricle, and hippocampus/amygdala complex. There were no differences in the tDCS-induced electric field among AD and normal groups; however, higher peak values of electric field were observed in the female group than the male group. Conclusions: Our findings demonstrated the presence of sex and disease differences in segmented brain volumes; however, this pattern differed in tDCS-induced electric field, resulting in significant sex differences only. Further studies, we will adjust the brain stimulation conditions to target the deep brain and examine the effects, because of significant differences in the ventricles and deep brain regions between AD and normal groups.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.81-88
• /
• 2023

Recently, communication through online is increasing due to the spread of non-face-to-face services due to COVID-19. In non-face-to-face situations, the other person's opinions and emotions are recognized through modalities such as text, speech, and images. Currently, research on multimodal emotion recognition that combines various modalities is actively underway. Among them, emotion recognition using speech data is attracting attention as a means of understanding emotions through sound and language information, but most of the time, emotions are recognized using a single speech feature value. However, because a variety of emotions exist in a complex manner in a conversation, a method for recognizing multiple emotions is needed. Therefore, in this paper, we propose a multi-emotion regression model that extracts feature vectors after preprocessing speech data to recognize complex, inherent emotions and takes into account the passage of time.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.6
• /
• pp.47-55
• /
• 2003

In this paper, we propose a new approach to detect peculiar features of license plates using intensity vector and composite color component in order to extract license plates from parking regulation images, which is captured in various locations around the front or the rear of cars at various times and places, and in which complex background is included. We fundamentally use both features that intensity value repeats frequently increasing and decreasing because intensity is obviously different at numerics and background, and that color is uniform in the area of license plates. First, we search each row at regular intervals starting from the bottom of a license-plate image, and we set up a rough region for a certain zone in which tile sign of intensity vector changes frequently enough and color of license plate is detected enough, assuming it as a candidate location of a license plate. And then, we extract an elaborate area of a license plate by projecting vertical edges horizontally and vertically. Here, type of cars, such as the urinate and the public, is easily classified according to the color of extracted plates. We used 200 actual regulation images, which are captured at various times and places, to evaluate the performance of the proposed method. As a result, the proposed method showed extraction rate of 96%, which is 9% higher than the previous method using only intensity vector.

• Archives of design research
• /
• v.17 no.2
• /
• pp.15-22
• /
• 2004

Recent complex visual culture is the visual world widely magnified according to the images like image, graphics, photograph, movie, and television, etc. by the development of digital technology. Because it conveys meanings and contents inserting sound and letters, it may have multimedia character conveying and communicating information beyond general language and letters. The vision for various images at that time is inseparably connected with language. And imaginative order of image and vision are composed of special way in culture and history. Language is different in society, culture, and history. Accordingly, if visual experience is communicated with language partially, it is difficult to have university. So, role of linguistic order plays an important role in forming and defining the social and cultural differences among the visual systems. Historically various visual and optical devices with this visual language have influenced a lot. These visual technologies are concrete and physical practice determining a way to get together with the subject and the visible object in the visible world. The visual language is connected with dimension like these symbols of images and the dimension like visual technologies to series of historical physical and institutional practices. It determines social visual mode toward object world in one of visual system. Accordingly, this study is to understand visual language with social and historical character according to the changed concept and characters as development of media technology. And it is to explain it in view of visual language as a dimension of symbol and visual technology of institutional and physical practice. After all, it cannot explain the effect on the function and visual mode of visual technology as its technical element only. It also cannot separate with the practice with coherent discourse and the physical and institutional practice. The possibility, technical element of technology contains, does not realize as it is but the effect is always communicated in the social veins and realized with a restriction.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.10
• /
• pp.633-639
• /
• 2018

High-resolution images using remote sensing (RS) is importance to secure for spatial classification depending on the characteristics of the complex and various factors that make up the river environment. The purpose of this study is to evaluate the accuracy of the classification results and to suggest the possibility of applying the high resolution hyperspectral images obtained by using the drone to perform spatial classification. Hyperspectral images obtained from study area were reduced the dimensionality with PCA and MNF transformation to remove effects of noise. Spatial classification was performed by supervised classifications such as MLC(Maximum Likelihood Classification), SVM(Support Vector Machine) and SAM(Spectral Angle Mapping). In overall, the highest classification accuracy was showed when the MLC supervised classification was used by MNF transformed image. However, it was confirmed that the misclassification was mainly found in the boundary of some classes including water body and the shadowing area. The results of this study can be used as basic data for remote sensing using drone and hyperspectral sensor, and it is expected that it can be applied to a wider range of river environments through the development of additional algorithms.

• Korean Journal of Remote Sensing
• /
• v.29 no.4
• /
• pp.361-373
• /
• 2013

The monitoring of urban area, target detection and building reconstruction have been actively studied and investigated since high resolution X-band SAR images could be acquired by airborne and/or satellite SAR systems. This paper describes an efficient approach to reconstruct artificial structures (e.g. apartment, building and house) in urban area using high resolution X-band SAR images. Building footprint was first extracted from 1:25,000 digital topographic map and then a corner line of building was detected by an automatic detecting algorithm. With SAR amplitude images, an initial building height was calculated by the length of layover estimated using KS-test (Kolmogorov-Smirnov test) from the corner line. The interferometric SAR phases were simulated depending on SAR geometry and changable building heights ranging from -10 m to +10 m of the initial building height. With an interferogram from real SAR data set, the simulation results were compared using the method of the phase consistency. One of results can be finally defined as the reconstructed building height. The developed algorithm was applied to repeat-pass TerraSAR-X spotlight mode data set over an apartment complex in Daejeon city, Korea. The final building heights were validated against reference heights extracted from LiDAR DSM, with an RMSE (Root Mean Square Error) of about 1~2m.