• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.1
• /
• pp.31-37
• /
• 2009

This paper casts an application of an extended fuzzy integral on the selective extraction of color clusters characterized by a particular color hue from color document images. Despite of the negative role of fuzzy integral, the presented approach attains the detection of the seals through the neighborhood information via the center of area method. The conventional fuzzy integral evaluates the negative aspects of the importance about the items by min operator, which result in the discontinuous parts of seals. In an attempt to cope with the drawback our approach considers the integral aspects via the center of area method, which results in the robustness of the images. Finally, the framework is successfully tested on a data set formed by documents from a real application for the detection.

• Journal of Korean Neurosurgical Society
• /
• v.54 no.1
• /
• pp.30-33
• /
• 2013

Objective : To suggest a new useful diagnostic technique, principles of the selective excitation technique-magnetic resonance images (Proset-MRI), and to know the precise radiologic findings that can prove symptomatic foraminal and extraforaminal stenosis at L5-S1. Methods : Nineteen patients with symptomatic L5-S1 stenosis were checked by Proset-MRI. Four patients were performed decompressive surgery and 15 patients were performed selective nerve root block (SNRB) at L5. The pain scale of patients was checked by Visual Analogue Scale (VAS) scores at the pre- and post-treatment state. Results : Proset-MRI findings of patients with symptomatic stenosis are root swelling (RS) and indentation. The comparisons with VAS scores had a meaningful statistical result at each RS (p<0.01) and indentation (p<0.01). However, the findings of RS combined with indentation lacked statistical significance (p=0.0249). In addition, according to a comparison with the treatment modalities, reducing of VAS scores had statistical meaningful significance in decompressive surgery cases (p<0.01), and also in SNRB cases (p<0.01) after a 3-month follow-up period. Conclusion : The three dimensional Proset-MRI is very useful and sensitive technique to diagnose the symptomatic foraminal and extraforaminal stenosis at L5-S1.

• Journal of KIISE:Software and Applications
• /
• v.32 no.1
• /
• pp.41-49
• /
• 2005

In this Paper, we Propose an automatic lung registration technique using local distance propagation for correcting the difference between two temporal images by a patient's movement in abdomen CT image obtained from the same patient to be taken at different time. The proposed method is composed of three steps. First, lung boundaries of two temporal volumes are extracted, and optimal bounding volumes including a lung are initially registered. Second, 3D distance map is generated from lung boundaries in the initially taken volume data by local distance propagation. Third, two images are registered where the distance between two surfaces is minimized by selective distance measure. In the experiment, we evaluate a speed and robustness using three patients' data by comparing chamfer-matching registration. Our proposed method shows that two volumes can be registered at optimal location rapidly. and robustly using selective distance measure on locally propagated 3D distance map.

• 한국노년학
• /
• v.29 no.3
• /
• pp.1001-1024
• /
• 2009

This study investigates what older adults have image of older adult with relation to their self-identity. Fourteen participants are interviewed in this study. The main findings are as follows. Their response to image of older adult is divided in the form of 'denial' and 'accommodation' about social images of older adults. 'Denial' has 'total denial' and 'selective denial' while 'accommodation' includes 'acceptive accommodation', 'selective accommodation' and 'unwilling accommodation'. 'Total denial' indicates that the participants describe older adults very negatively, and they regard themselves as a person who is totally different from the older person they describe. 'Selective denial' represents that the participants have two types of older adults: negative and positive. They describe that they are close to the positive type of older adults themselves although they do not admit that they are older adults. 'Acceptive accommodating' does not describe older adults negatively, and they identify themselves as older adults. 'Selective accommodating' are very similar to 'selective denial' although the former differs from the latter in that they accept that they are older adults. 'Selective accommodating' also includes negative and positive types of older adults, and this type of participants admit that they are older adults, and believe that they belong to the positive type. In contrast to 'acceptive accommodating', 'unwilling accommodating' describes older adults very negatively. However, the participants belonging to 'unwilling accommodating' comes to accept that they are older adults due to irresistible realities.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.21-32
• /
• 2022

Image labeling must be preceded in order to perform object detection, and this task is considered a significant burden in building a deep learning model. Tens of thousands of images need to be trained for building a deep learning model, and human labelers have many limitations in labeling these images manually. In order to overcome these difficulties, this study proposes a method to perform object detection without significant performance degradation, even though labeling some images rather than the entire image. Specifically, in this study, low-resolution oriental painting images are converted into high-quality images using a super-resolution algorithm, and the effect of SSIM and PSNR derived in this process on the mAP of object detection is analyzed. We expect that the results of this study can contribute significantly to constructing deep learning models such as image classification, object detection, and image segmentation that require efficient image labeling.

• Journal of Digital Contents Society
• /
• v.19 no.4
• /
• pp.757-770
• /
• 2018

Foreground extraction is the most significant step in thermal imaging based surveillance systems. This step needs to be efficient in terms of time and memory consumption in order for the system to provide real time results but usually this efficiency reciprocates with the accurateness of the ROI detection. In this study, novel selective histogram bins based two background & foreground separation approaches for thermal videos processing have been proposed which exploit the temporal-consistency property of the thermal images in a given environment and can save over 80% memory than their simplest counterpart temporal median filtering.

• Journal of Environmental Science International
• /
• v.29 no.1
• /
• pp.95-108
• /
• 2020

Silver nanoparticles were loaded onto g-C₃N₄ (CN) with a nanoroll-type morphology (Ag/CN) synthesized using a co-polymerization method for highly selective conversion of toxic nitrobenzene to industrially-valuable aminobenzene. Scanning electron microscopy and high-resolution transmission electron microscopy (HRTEM) images of Ag/CN revealed the generation of the nanoroll-type morphology of CN. Additionally, HRTEM analysis provided direct evidence of the generation of a Schottky barrier between Ag and CN in the Ag/CN nanohybrid. Photoluminescence analysis and photocurrent measurements suggested that the introduction of Ag into CN could minimize charge recombination rates, enhancing the mobility of electrons and holes to the surface of the photocatalyst. Compared to pristine CN, Ag/CN displayed much higher ability in the photocatalytic reduction of nitrobenzene to aminobenzene, underscoring the importance of Ag deposition on CN. The enhanced photocatalytic performance and photocurrent generation were primarily ascribed to the Schottky junction formed at the Ag/CN interface, greater visible-light absorption efficiency, and improved charge separation associated with the nanoroll morphology of CN. Ag would act as an electron sink/trapping center, enhancing the charge separation, and also serve as a good co-catalyst. Overall, the synergistic effects of these features of Ag/CN improved the photocatalytic conversion of nitrobenzene to aminobenzene.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.1299-1302
• /
• 2004

A real object duplication system (RODS), including three dimensional (3D) scanner and solid freeform fabrication system (SFFS), is a device to make three-dimensional objects directly from the drawing or photo data. A Selective Multi-Laser Sintering (SMLS) process designed in this paper is by which computer images received using 3D scanner are built up from polymer powder on building room of large size using dual laser at industrial type SFF system. Using the process can rapidly produce real object duplication components of industrial type such as cylinder, engine block, chassis of automobile, etc. In this paper, the industrial type SFF system using SMLS process is manufactured and the system is satisfied with high precision and high speed processing technique. To research characteristics of each part for theindustrial type SFF system, a structure and thermal analysis and test of each part is carried out. Also, to achievement of high performance for industrial type SFF system, design and fabrication for the structure, heater, nitrogen supply, laser and control part are carried out.

• Journal of Korea Multimedia Society
• /
• v.18 no.11
• /
• pp.1289-1301
• /
• 2015

Pedestrian detection (PD) is an essential task in various applications and sliding window-based methods utilizing HOG (Histogram of Oriented Gradients) or HOG-like descriptors have been shown to be very effective for accurate PD. However, due to exhaustive search across images, PD methods based on sliding window usually require heavy computational time. In this paper, we propose a real-time PD method for embedded visual surveillance with fixed backgrounds. The proposed PD method employs HOG descriptors as many PD methods does, but utilizes selective search so that it can save processing time significantly. The proposed selective search is guided by restricting searching to candidate regions extracted from Adaptive Gaussian Mixture Model (AGMM)-based background subtraction technique. Moreover, approximate computation of HOG descriptor and implementation in fixed-point arithmetic mode contributes to reduction of processing time further. Possible accuracy degradation due to approximate computation is compensated by applying an appropriate one among three offline trained SVM classifiers according to sizes of candidate regions. The experimental results show that the proposed PD method significantly improves processing speed without noticeable accuracy degradation compared to the original HOG-based PD and HOG with cascade SVM so that it is a suitable real-time PD implementation for embedded surveillance systems.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.12 no.4
• /
• pp.231-238
• /
• 2017

Detecting small infrared targets from the low-SCR images at a long distance is very hard. The previous Local Contrast Method (LCM) algorithm based on the human visual system shows a superior performance of detecting small targets by a background suppression technique through local contrast measure. However, its slow processing speed due to the heavy multi-scale processing overhead is not suitable to a variety of real-time applications. This paper presents a lightweight real-time small target detection algorithm, called by the Improved Selective Local Contrast Method (ISLCM), to reduce the scale-invariant computational overhead. The proposed ISLCM applies the improved local contrast measure to the predicted selective region so that it may have a comparable detection performance as the previous LCM while guaranteeing low scale-invariant computational load by exploiting both adaptive scale estimation and small target feature feasibility. Experimental results show that the proposed algorithm can reduce its computational overhead considerably while maintaining its detection performance compared with the previous LCM.