• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.404-406
• /
• 2021

In this paper, we propose an app application for classifying Korean food images and providing information related to Korean food. App Application consists of Flask server, Database (Mysql), and Python deep learning modules. Using the VGG16 model, 150 images of Korean foods are classified. If there is an internet environment, anyone can easily get information about Korean food anytime, anywhere with a single photo.

• Journal of the Society of Disaster Information
• /
• v.14 no.4
• /
• pp.501-509
• /
• 2018

Purpose: UAV-based photo measurements are being researched using UAVs in the space information field as they are not only cost-effective compared to conventional aerial imaging but also easy to obtain high-resolution data on desired time and location. In this study, the UAV-based high-resolution images were used to perform the land cover classification. Method: RGB cameras were used to obtain high-resolution images, and in addition, multi-distribution cameras were used to photograph the same regions in order to accurately classify the feeding areas. Finally, Land cover classification was carried out for a total of seven classes using created ortho image by RGB and multispectral camera, DSM(Digital Surface Model), NDVI(Normalized Difference Vegetation Index), GLCM(Gray-Level Co-occurrence Matrix) using RF (Random Forest), a representative supervisory classification system. Results: To assess the accuracy of the classification, an accuracy assessment based on the error matrix was conducted, and the accuracy assessment results were verified that the proposed method could effectively classify classes in the region by comparing with the supervisory results using RGB images only. Conclusion: In case of adding orthoimage, multispectral image, NDVI and GLCM proposed in this study, accuracy was higher than that of conventional orthoimage. Future research will attempt to improve classification accuracy through the development of additional input data.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.20 no.3
• /
• pp.63-78
• /
• 2018

This study analyzed gender-fluid images as expressions of feminism and gender identity expressed in fashion collections. As for the research method, this study searched the fashion collections, with the theme of feminism, utilizing key words related to feminism on an online portal, and collected the photo materials of fashion collections provided by vogue.com. This study classified the photo materials of 31 fashion collections, with the theme of feminism, into femininity, masculinity, androgyny, and avant-garde, according to the fashion design elements that divide gender identity. As a result of the classification, 326 photos were collected, in which gender identity was expressed ambiguously. This study reclassified the collected photos according to their fashion items and styles. As a result of the study, it was noticed that the fashion collections with the theme of feminism expressed the messages, using lettering graphic images, and performance. In addition, they showed a form in which men's collections and women's collections were integrated according to the change of the perceptions of gender identity, of feminism, and delivered body positive expressions, respecting differences and diversity as individual subjects, by casting diverse models in terms of age, body size, race, and culture. As for the gender identity expressed in the fashion collections, the gender-fluid images were classified into empowerment images, that expresses social rights and dignity; agender images that expresses the possibility of a gender-flexible transition; rational images that expresses the rational and practical characteristics that removed the boundary of fashion; and images of pro-sexism that expresses a new gender identity.

• Korean Journal of Remote Sensing
• /
• v.35 no.2
• /
• pp.229-239
• /
• 2019

All vegetation colonies have layered structure. This layer is called 'forest vertical structure.' Nowadays it is considered as an important indicator to estimate forest's vital condition, diversity and environmental effect of forest. So forest vertical structure should be surveyed. However, vertical structure is a kind of inner structure, so forest surveys are generally conducted through field surveys, a traditional forest inventory method which costs plenty of time and budget. Therefore, in this study, we propose a useful method to classify the vertical structure of forests using remote sensing aerial photographs and machine learning capable of mass data mining in order to reduce time and budget for forest vertical structure investigation. We classified it as SVM (Support Vector Machine) using RGB airborne photos and LiDAR (Light Detection and Ranging) DSM (Digital Surface Model) DTM (Digital Terrain Model). Accuracy based on pixel count is 66.22% when compared to field survey results. It is concluded that classification accuracy of layer classification is relatively high for single-layer and multi-layer classification, but it was concluded that it is difficult in multi-layer classification. The results of this study are expected to further develop the field of machine learning research on vegetation structure by collecting various vegetation data and image data in the future.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.32-38
• /
• 2021

Image classification through deep learning on the image from photographs has been a very active research field for the past several years. In this paper, we propose a method of automatically discriminating stone images from domestic source through deep learning, which is to use Python's hash library to scan 300×300 pixel photo images of granites such as Hwangdeungseok, Goheungseok, and Pocheonseok, performing data preprocessing to create learning images by examining duplicate images for each stone, removing duplicate images with the same hash value as a result of the inspection, and deep learning by stone. In addition, to utilize VGGNet, the size of the images for each stone is resized to 224×224 pixels, learned in VGG16 where the ratio of training and verification data for learning is 80% versus 20%. After training of deep learning, the loss function graph and the accuracy graph were generated, and the prediction results of the deep learning model were output for the three kinds of stone images.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.5
• /
• pp.535-546
• /
• 2020

Land cover maps are derived from satellite and aerial images by the Ministry of Environment for the entire Korea since 1998. Even with their wide application in many sectors, their usage in research community is limited. The main reason for this is the map compilation cycle varies too much over the different regions. The situation requires us a new and quicker methodology for generating land cover maps. This study was conducted to automatically generate land cover map using aerial ortho-images and Landsat 8 satellite images. The input aerial and Landsat 8 image data were trained by Residual U-Net, one of the deep learning-based segmentation techniques. Study was carried out by dividing three groups. First and second group include part of level-II (medium) categories and third uses group level-III (large) classification category defined in land cover map. In the first group, the results using all 7 classes showed 86.6 % of classification accuracy The other two groups, which include level-II class, showed 71 % of classification accuracy. Based on the results of the study, the deep learning-based research for generating automatic level-III classification was presented.

• Korean Journal of Remote Sensing
• /
• v.22 no.6
• /
• pp.565-574
• /
• 2006

Satellite images have been used to obtain land cover information that is one of important factors for hydrological analysis over a large area. In urban area, more detailed land cover data are often required for hydrological analysis because of the relatively complex land cover types. The number of land cover classes that can be classified with traditional multispectral data is usually less than the ones required by most hydrological uses. In this study, we present the capabilities of hyperspectral data (Hyperion) for the classification of hydrological land cover types in urban area. To obtain 17 classes of urban land cover defined by the USDA SCS, spectral mixture analysis was applied using eight endmembers representing both impervious and pervious surfaces. Fractional values from the spectral mixture analysis were then reclassified into 17 cover types according to the ratio of impervious and pervious materials. The classification accuracy was then assessed by aerial photo interpretation over 10 sample plots.

• Architectural research
• /
• v.12 no.2
• /
• pp.53-62
• /
• 2010

The purpose of this study is to examine design preferences for the sales spaces of duty-free shops (DFSs) by conducting image evaluations. The results will help improve quality by influencing designs for the construction, extension or remodeling of these shops. An image measurement method, the semantic differential method, was used to measure cognitive structure using photos of shops. Photos were collected of the DFS at Jeju Island, as well as photos of brand stores designed by architects. Two sets of 16 photos (32 different photos in all) were selected according to photo classification standards and design concepts, both decided by reviewing previous studies and related materials. The evaluation and survey were done by two sets of subjects: sales employees, who have experience and special knowledge of the evaluation of sales space; and students majoring in architecture. To strengthen the evaluation results, I conducted a preliminary survey and a main survey, verifying and complementing findings. 116 surveys were conducted, of which 14 were of poor quality and rejected, leaving and 102 to be analyzed. The collected surveys were statistically analyzed, using SPSS 12.0 for Windows. Reliability, image profile, factor and multi-dimensional scaling analyses were conducted. As a result, image evaluation structure and characteristics were obtained for sales spaces of DFSs, confirming the difference between them and other spaces.

• Particle and aerosol research
• /
• v.13 no.1
• /
• pp.25-31
• /
• 2017

This paper offer the characteristics for the detection and classification of biological and non-biological aerosol particles in the air by using laser-induced-fluorescence (LIF) based Bio-aerosol Detection System (BDS). The BDS is mainly consist of an optical chamber, in-outlet nozzle system, 405 nm diode laser, an avalanche photo detector (APD) for scattering signal and photomultiplier tubes (PMT) for fluorescence signals in two different wavelength range ; F1, 510-600 nm and F2, 435-470 nm. The detection characteristics, especially ratio of fluorescence signal intensity were examined using well-known components : polystylene latex (PSL), fluorescence PSL, $2{\mu}m$ of SiO2 micro sphere, dried yeast, NADH, ovalbumin, fungicide powder and standard dust. The results indicated that the 405 nm diode laser-based LIF instrument can be a useful bio-aerosol detection system for unexpected biological threaten alter in real-time to apply for dual-use technology in military and civilian fields.

• Journal of Internet Computing and Services
• /
• v.5 no.5
• /
• pp.79-91
• /
• 2004

This paper proposes avatar face generation system that uses shading mechanism and facial features extraction method of facial recognition. Proposed system generates avatar face similar to human face automatically using facial features that extracted from a photo. And proposed system is an approach which compose shade and facial features. Thus, it has advantages that can make more realistic avatar face similar to human face. This paper proposes new eye localization method, facial features extraction method, classification method for minimizing retrieval time, image retrieval method by similarity measure, and realistic avatar face generation method by mapping facial features with shaded face pane.