• The Journal of Korea Robotics Society
• /
• v.16 no.4
• /
• pp.299-305
• /
• 2021

In this study, the task of robotic tidy-up is to clean the current environment up exactly like a target image. To perform a tidy-up task using a robot, it is necessary to estimate the pose of various objects and to classify the objects. Pose estimation requires the CAD model of an object, but these models of most objects in daily life are not available. Therefore, this study proposes an algorithm that uses point cloud and PCA to estimate the pose of objects without the help of CAD models in cluttered environments. In addition, objects are usually detected using a deep learning-based object detection. However, this method has a limitation in that only the learned objects can be recognized, and it may take a long time to learn. This study proposes an image matching based on few-shot learning and Siamese network. It was shown from experiments that the proposed method can be effectively applied to the robotic tidy-up system, which showed a success rate of 85% in the tidy-up task.

• Archives of design research
• /
• v.15 no.1
• /
• pp.359-368
• /
• 2002

Fabric art is most distinguished field among the Modern art. Among them, tapestry which begins human history is investigated more and more deeply and till now it displays with various type and exhibition. But, because of the huge scale of the working process, it requires much time and hard endeavor. After sampling, it begins main process in the present situation. But we cannot know exactly whether it goes well or not until it ends. So after fulfilled whole process, we often try new work again. Because of this reason, we devise computer simulation. With computer simulation we can preview whole working and lengthen the planning time, shorten the needless time. Simulation is made up of Scan, Drawing, Line clean up, Rendering, Parity, and Printing. Scan and Drawing can male new idea. And during the Line clean up we can make vector image and modify the image. And then rendering the image and inquiring the length and weight of yarn. Last process is printing an then making a package soft ware by means of prototype. Through these process, many producers and student can easily access to tapestry and reduce the needless time and endeavor.

• Proceedings of the Korean Institute of Interior Design Conference
• /
• 1999.04a
• /
• pp.87-93
• /
• 1999

In the millenium, which will be characterized by its global culture, a strong sense of identity of one's culture is needed to nature one's culture in a way that more people can participate in creating, enjoying and consuming it. In order to respond to both globalization and localization, natural and cultural backgrounds of such should be considered in the conversation of nature, so that the optical and color environment can have local identity while undergoing sustainable development. Therefore, the objective of this research is to enhance the image of a city, which is endowed with beautiful scenery and unique culture, but has lost its identity since industrialization An through a scientific and systematic color planning The Research was conducted as follows. First, color environment which doesn't reflect local characteristics due to modernism which emphasizes on uniformity was examined. Second, standard and analysis of environment color planning were examined. Natural color specimen was use to make a palette of nature colors, from which colors for architecture were extracted to come up with color combination types. Without traditional architecture philosophy, materials and colors, which were once in harmony with the environment, it is impossible to clean up the environment all together. However by using colors to control new city plans, urban redevelpment, construction and renovation gradually, that process can be effective and less costly.

• Journal of Korea Multimedia Society
• /
• v.23 no.11
• /
• pp.1349-1360
• /
• 2020

In recent years, Convolutional Neural Networks (CNNs) have been successfully implemented in different tasks of computer vision. Since CNN models are the representatives of supervised learning algorithms, they demand large amount of data in order to train the classifiers. Thus, obtaining data with correct labels is imperative to attain the state-of-the-art performance of the CNN models. However, labelling datasets is quite tedious and expensive process, therefore real-life datasets often exhibit incorrect labels. Although the issue of poorly labelled datasets has been studied before, we have noticed that the methods are very complex and hard to reproduce. Therefore, in this research work, we propose Deep CleanNet - a considerably simple system that achieves competitive results when compared to the existing methods. We use K-means clustering algorithm for selecting data with correct labels and train the new dataset using a deep CNN model. The technique achieves competitive results in both training and validation stages. We conducted experiments using MNIST database of handwritten digits with 50% corrupted labels and achieved up to 10 and 20% increase in training and validation sets accuracy scores, respectively.

• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.369-376
• /
• 2006

The degradation and its inverse modeling can achieve restoration of corrupted image, caused by scaled digitization and electronic transmission. De-speckle process on the noisy document(or SAR) images is one of the basic examples. Non-linearity of the speckle noise model may hinder the inverse process. In this paper, our study is focused on investigation of the restoration methods for bi-level document image degraded by the impulse noise model. Our study shows that, on bi-level document images, the weighted-median filter and the Lee filter methods are very effective among other spatial filtering methods, but wavelet filter method is ineffective in aspect of processing speed: approximately 100 times slower. Optimal values of the weight to be used in the weighted median filter are investigated and presented in this paper.

• Journal of Korean Society of Rural Planning
• /
• v.21 no.2
• /
• pp.163-175
• /
• 2015

This study aimed to analyze factors of importance and performance picked by customers of Bibijeong, a farm restaurant operated by the local community of Wanju-gun. Major points include: Recognition paths mostly were word of mouth and mass media, types of visit were in the company of friends, family members or work colleagues, and purposes of visit included consumption of meal and identifying features of the restaurant. Secondly, factor analysis showed that level of facility, atmosphere/cleanness, diversity of menu, employees, ingredients and network. The Cronbach Alpha coefficient was +0.6. Thirdly, average of importance of factors was 3.861 while average performance was 3.429. IPA analysis showed that employee(communication, customer contact) in the first quadrant proved the need for fast improvement through training. Atmosphere/cleanness (interior atmosphere, table clean, kitchen cleanliness, clean dishes, interion design) and employee(proficiency, menu recognition), foodstuff(freshness, origin, safety) in the second quadrant showed that the marketing strategy of improvement as well as maintaining current status is needed, including regular training and hygiene inspection. The third quadrant contains facilities(disability, baby, fire protection) and food menu(food packing, various menu, creative menu, menu description), network(village economic links), which showed the need for gradual improvement. The forth quadrant contains network(sights's near contains. The results so far can be summed into the statement that overcoming the basic functionality of providing meals and linking the restaurant with local attractions and local economy would be need, as well as building up the image of unique farm restaurant with local features, so that Bibijeong can serve as the centerpiece of community and foundation of exchange with other areas.

• Journal of the Korean Society for Precision Engineering
• /
• v.29 no.5
• /
• pp.545-553
• /
• 2012

A Gimbal structure system in observation reconnaissance aircraft is made up of camera module and stabilization drive device supporting camera module. During flight for image recording, the aircraft undergoes serious accelerations with wide frequencies due to several factors. Though base excitation of stabilization drive device induces vibration of camera module, it must get the stable and clean images. To achieve this aim, acceleration of camera module must be reduced. Hence, vibration isolators were installed to stabilization drive device. Considering isolators and bearings in the stabilization drive device, vibration characteristics of gimbal structure system were analyzed by finite element method. For three translational direction, acceleration transmissibility of camera module was calculated by harmonic responses analysis in the frequency range of 5 ~ 500 Hz. In addition to, sine-sweep experiment were performed to prove correctness of present analysis.

• Clean Technology
• /
• v.15 no.4
• /
• pp.265-272
• /
• 2009

In this study the experiments on the static adsorption of benzene were carried out using activated carbon made from sewage sludge. The experiment was performed at 303.15 K, 318.15 K and 333.15 K under the pressure up to 7.999kPa. Isothermal adsorption curves were obtained using Langmuir isotherm, Freundlich isotherm and Toth isotherm for comparison. Based on fitting the adsorption quantity of Benzene (q), the isothermal adsorption curves obtained from Langmuir isotherm and Toth isotherm showed the higher accuracy. Although there was little difference in accuracy between result from Langmuir isotherm and that from Toth isotherm, the adsorption quantity of Benzene (q) was expressed in terms of Langmuir isotherm because less parameters were required for Langmuir isotherm than for Toth isotherm. Moreover SEM images of the activated carbon from sewage sludge and the commercial activated carbon were taken to observe the pore size development. The results showed that the perforation development of the commercial activated carbon (DARCO A.C., SPG-100 A.C.) was better than that of activated carbon from sewage sludge. Adsorption quantity of benzene on commercial activated carbon was confirmed to be higher than that on activated carbon from sewage sludge. However the maximum adsorption quantity of benzene on activated carbon from sewage sludge was close to that on SGP-100 A.C. at 303.15K. Therefore, we may conclude that it is feasible to commercialize the process to manufacturing activated carbon from sewage sludge.

• Journal of the Korean GEO-environmental Society
• /
• v.10 no.6
• /
• pp.49-60
• /
• 2009

In order to develop a new rockfall protection fence using high carbon steel wire rod (HSWR) material instead of the conventional wire rope material, the author has conducted the laboratory strength tests of both materials and their connections, and carried out evaluation of absorbing rockfall energy through the vertical field rockfall tests. The vertical filed rockfall tests showed that the new rockfall protection fence with 12 rows of the HSWR could absorb more rockfall energy than 50 kJ which stands for the typical design criteria. In addition, when the quantity of HSWR was increased up to the 16 rows, the capacity of absorbing energy was greatly improved. The new rockfall protection fence was successfully applied to the highway rock-cut slope. As a result of the filed application, its constructability was similar to the conventional fence, but its total image was improved as simple and clean. The total construction cost was saved up to 20% in comparison with the conventional one.

• Journal of Acupuncture Research
• /
• v.31 no.1
• /
• pp.75-93
• /
• 2014

Objectives : This study was designed to establish medical trust realization system by finding factors influential to it using questionnaire. Methods : 277 subjects were participated in this study. After a treatment, we conducted a survey from April 1st to October 31th about medical service perception index, medical trust index, patient satisfaction index, patient reliability index, patient flow degree index, recall intension index, and hospital image index. To evaluate the influence of medical service perception with other 6 indexes, we statistically made regression analysis of the results through the survey. Results : By the results of the analysis, evaluation of hospital image influenced all 6 indexes. The systemicity of treatment process had an effect on 5 indexes except for the flow degree of patients. The humanity of medical team brought out the estimation of 4 indexes except for the patient flow degree and hospital image. The empathic ability of doctor and appropriacy of medical costs hold the next rank influencing 3 indexes. It reached the conclusion that the systemicity of medical team tend to determine the medical trust and patient reliability. The expertise, professional skill of doctor, the fault, commercial application, fame of medical team, the speed of treatment process, the newest and clean medical facility affected each one index. Conclusions : Korean medicine should find a way to consider the mind of patients for improving the medical quality through trust realization system, keeping up with times. As a result of this research, we can find out important causes which influence the trustful medical system. From now on, we should apply this result to actual treatment of psychology customized system. Also, more simple and clear questionnaire was organized through this research, it can be used to forward research to apprehend patient mentality more conveniently.