• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.99-101
• /
• 2021

쓰레기 대란, 환경파괴의 상황 속 실제 재활용 쓰레기 가운데 절반 정도만이 재활용되고 있다. 재활용률을 높이기 위해, 올바른 분리배출 방법을 쉽고 편하게 찾을 수 있는 방식이 필요하다. 본 논문에서는 올바른 분리수거를 통해 재활용률을 증진하기 위한 분리수거 분류 서비스를 제안한다. 본 논문은 ResNet-34 모델을 통해 안드로이드 카메라로 촬영한 이미지의 분리배출 클래스를 예측하고 그에 따른 분리배출 가이드를 제공하는 시스템을 설계하였다. 향후 연구에서는 모델의 정확도 향상을 위해 온디바이스와 서버 모델을 분리하고 모델의 개인 맞춤화를 진행할 예정이다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.5-6
• /
• 2020

현대 의학에서 의료 영상은 수많은 영상처리 의료기기의 핵심이다. PACS(Picture Archiving Communication System)를 통해 관리되는 의료 영상 자료들은 요청에 따라 저장, 검색 및 전송을 수행하여 신속한 의료 서비스를 가능하게 한다. 그러나 만약에 관리자의 실수로 의료 영상 데이터가 바뀐다면 이는 사용자로 하여금 불편함과 낮은 신뢰성을 야기한다. 그리하여 본 논문에서는 서포트 벡터 머신 기반의 HOG(Histogram of Oriented Gradients) 특징 벡터를 이용하여 X-ray와 MRI(Magnetic Resonance Imaging) 사진을 분류하고 의료 영상 분류의 가능성을 제시하는 것을 목표로 한다.

• Journal of The Korean Association For Science Education
• /
• v.31 no.5
• /
• pp.709-719
• /
• 2011

The purpose of this study was to identify the differences in classification skills shown in classification activities between the gifted and regular students in elementary schools. The subjects for the research consisted of six gifted students in an institute for the gifted for science annexed to P school district in Gangwon-do and 6 students at B and M general elementary schools. Results were as follows: The time taken for classification activities of the gifted was shorter than regular regardless of subjects for classifying. The number of standards for classifying for the gifted was more than regular students. Coefficient for measuring classification skills of the gifted was higher than regulars regardless of age. Consequently, there was a difference in the time taken for classifying and generating the number of standards and in a numerical index of classification activities performed at science classes between the science gifted and the regular students.

• Journal of Korea Multimedia Society
• /
• v.4 no.2
• /
• pp.171-181
• /
• 2001

The previous Bayesian document categorization method has problems that it requires a lot of time and effort in word clustering and it hardly reflects the semantic information between words. In this paper, we propose a weighted Bayesian document categorizing method based on association word knowledge base acquired by mining technique. The proposed method constructs weighted association word knowledge base using documents in training set. Then, classifier using Bayesian probability categorizes documents based on the constructed association word knowledge base. In order to evaluate performance of the proposed method, we compare our experimental results with those of weighted Bayesian document categorizing method using vocabulary dictionary by mutual information, weighted Bayesian document categorizing method, and simple Bayesian document categorizing method. The experimental result shows that weighted Bayesian categorizing method using association word knowledge base has improved performance 0.87% and 2.77% and 5.09% over weighted Bayesian categorizing method using vocabulary dictionary by mutual information and weighted Bayesian method and simple Bayesian method, respectively.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.435-444
• /
• 2019

Recently, DL (Deep Learning) has been rapidly applied in various fields. In particular, classification and object recognition from images are major tasks in computer vision. Most of the DL utilizing imagery is primarily based on the CNN (Convolutional Neural Network) and improving performance of the DL model is main issue. While most CNNs are involve with images for training data, this paper aims to classify and recognize objects using DSM (Digital Surface Model), and slope and aspect information derived from the DSM instead of images. The DSM data sets used in the experiment were established by DGPF (German Society for Photogrammetry, Remote Sensing and Geoinformatics) and provided by ISPRS (International Society for Photogrammetry and Remote Sensing). The CNN-based SegNet model, that is evaluated as having excellent efficiency and performance, was used to train the data sets. In addition, this paper proposed a scheme for training data generation efficiently from the limited number of data. The results demonstrated DSM and derived data could be feasible for semantic classification with desirable accuracy using DL.

• Journal of Korean Society for Geospatial Information Science
• /
• v.18 no.1
• /
• pp.135-143
• /
• 2010

The national base map has its value sand roles as the basic spatial information of the nation. The current national base map that is a 1/5,000 digital map, however, has failed to perform its roles as basic spatial information due to the limitations with its quality and accuracy and requires measures to complement them. Thus this study set out to suggest ways to advance the current 1/5,000 national base map, selected topography and natural features of a digital map that could be made with GSD 0.25m digital aerial photographs, and set up the optimal ways to make a digital map by conducting an experiment of making an optimal digital map with such photographs. It also analyzed the map made with GSD 0.25m digital aerial photographs for accuracy and usability. In order to establish a set of criteria of making a digital map with GSD 0.25m digital aerial photographs, the investigator carried out analyses and picked topography and natural features items, which include 9 large categories, 31 medium categories, and 509 small categories. Then an experiment of making a digital map was conducted according to the digital map making method. As a result, solid drawing was selected as the optimal way to making a digital map, and the optimal process was established. Using the research achievements, a model digital map was made with GSD 0.25mm digital aerial photographs. The map recorded about two times horizontal and vertical location accuracy than the old 1/5,000 digital map and was capable of detailed descriptions of topography and natural features. A new national base map made with GSD 0.25m digital aerial photographs will provide reliable spatial data, thus increasing the level of satisfaction among people and the level of advancement of national base maps.

• Korean Journal of Human Ecology
• /
• v.12 no.5
• /
• pp.767-776
• /
• 2003

The purpose of this study was to classify side somatotypes of the trunk by analysing photographic data. Then their distribution according to the age groups was studied. The subjects were 315 females of 18 to 49 year-old. Thirty one photographic measurements were taken to each subject. The factors affecting the side somatotype of the trunk were obtained by principal component analysis, vertical size, posterior/anterior depth and neck posture. The side somatotypes of the trunk were classified into 4 types and their differences were shown by analysing photographic data. The side silhouettes of 4 types were compared with balanced type. By suggesting the canonical discriminant function with the unstandardized canonical coefficient, individual somatotype of the trunk could be discriminated from the photographic data of anterior neck height, anterior waist height, posterior waist depth, buttock height, and anterior depth at the level of back protrusion. The frequency distribution of the side somatotypes of the trunk according to the age groups could be applied for clothing construction and the rate of clothing production.

• Proceedings of the KSRS Conference
• /
• 2006.03a
• /
• pp.228-231
• /
• 2006

유출계수(Runoff Curve Number, CN)란 강수량으로부터 대상유역의 유출량과 우수 잠재능(stormwater potential) 평가에 이용하는 수문학 변수로, 미국 자연자원 보존국(Natural Resources Conservation Service; NRCS)이 제안한 방법이다. 유출계수를 평가하기 위해서는 토지피복, 토양형, 토양 습윤 조건에 대한 정보를 조합하여 분석해야 한다. 본 연구의 목적은 미국 North Carolina의 Raleigh와 Cary시를 관통하는 Walnut Creek 유역 서부지역의 토지 피복도를 제작하여, 이 유역의 유출계수를 산정하는 것이다. 이를 위해서, 첫째 위의 불투수면 레이어와 정사항공사진을 기초자료로, ArcGIS와 Feature Analyst를 이용하여 서부 Walnut Creek 유역의 토지피복도를 제작하였다. 둘째, 제작된 토지 피복도와 본 유역의 수문학적 토양 분류체계도(Hydrologic Soil Group Map)를 중첩하여 이 유역의 유출계수도를 제작하였다.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.1000-1005
• /
• 2007

본 논문에서는 실물체의 3차원 모델을 복원하기 위해 거리영상 카메라에서 획득된 3차원 점군에 대한 온라인 정합 기법을 제안한다. 제안하는 방법은 거리영상 카메라를 사용하여 연속된 거리영상과 사진영상을 획득하고 문턱값(threshold)을 이용하여 물체와 배경에 대한 정보를 분류한다. 거리영상에서 특징점을 선택하고 특징점에 해당하는 거리영상의 3차원 점군을 이용하여 투영 기반 정합을 실시한다. 초기정합이 종료되면 사진영상간의 대응점을 추적하여 거리영상을 정제하는 과정을 거치는데 대응점 추적에 사용되는 KLT(Kanade-Lucas-Tomasi) 추적기를 수정하여 초기정합의 결과를 대응점 탐색에 이용함으로써 탐색의 속도와 성공률을 증가시켰다. 특징점과 추적된 대응점에 해당하는 3차원 점군을 이용하여 거리영상의 정제를 수행하고 정합이 완료되면 오프라인에서 3차원 모델을 합성하였다. 제안한 알고리듬을 적용하여 2개의 실물체에 대하여 실험을 수행하고 3차원 모델을 생성하였다.

• Korean journal of applied entomology
• /
• v.57 no.3
• /
• pp.161-164
• /
• 2018

Pristiphora (Pristiphora) apricoti Zinovjev, 1993, pest of Prunus armeniaca var. ansu Max, belonging to Nematinae of Tenthredinidae is newly recognized in South Korea. The host plant is recorded for the first time from South Korea. Diagnosis, rearing notes, and photographs of the diagnostic characters and oviposition are provided.