• The Transactions of the Korea Information Processing Society
• /
• v.5 no.3
• /
• pp.833-841
• /
• 1998

The ann of this study is 10 propose all eff'tcient mclhod for partition of multi-dimensIOnal feature space into pattern subspace for automated generation of fuzzy rule. The suggested mclhod predicates on sequential subdivision of the fuzzy subspacc. and the size of construc1cd pattern space is variable. Under this procedure, n-dimensional pattern space, after considering the distributional characteristic patterns, is partitioned into two different pattern subspaces. From the two subspaces, the pattern space for further subdivision is chosen; then, this subdivision procedure recursively repeats itself until the stopping condition is fulfilled. The result of this study is applied to 2, 4, 7 band of satellite Landsat TM and satisfac10ry result is acquired.

• Annual Conference of KIPS
• /
• 2003.05a
• /
• pp.675-678
• /
• 2003

현대 사회에서 가스 산업은 기반 기순 산업으로 영향력이 매우 커지고 있으며, 동시에 위험요소를 내포하고 있다. 본 논문은 가스밸브의 연결부분에서 폭발성 발화물질인 오일성분에 의한 안전사고를 방지하고자 유지분 검출시스템을 개발하였다. 용기 밸브의 영상을 $200{\times}200$ 크기의 흑백 이미지로 획득하고, 제안한 유지분 검출 알고리즘을 적용한다. 먼저, 영상에 소벨(Sobel) 연산자를 이용하여 에지(Edge)를 검출하고 오일이 존재하는 영상과 오일이 존재하지 않는 영상을 영상프로파일(image profile) 분석과 영상 감법 둥의 검출방법으로 비교 분석하였다. 첫 번째 방법은 영상프로파일기법으로 3가지 영역인 부분영역, 하위영역, 전체영역으로 오일을 삽입한 후 다항식 근사로 오일의 존재여부를 판별하였고, 두 번째는 영상감법으로 밸브의 특정한 위치에 위치조정표시를 한 후 두 영상의 차를 구하였다. 실험에서는 제안한 검출방법의 실용화 가능성을 확인하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.1
• /
• pp.43-48
• /
• 2013

Adaboost is widely used for Haar-like feature boosting algorithm in Face Detection. It shows very effective performance on single distribution model. But when detecting front and side face images at same time, Adaboost shows it's limitation on multiple distribution data because it uses linear combination of basic classifier. This paper suggest the HDCT, modified decision tree algorithm for Haar-like features. We still tested the performance of HDCT compared with Adaboost on multiple distributed image recognition.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.429-431
• /
• 2019

Noise in image processing has a direct effect on the quality of the image, and adversely affects the processing of the system including algorithms such as image segmentation, edge detection, and image recognition. Therefore, noise reduction plays an important role in the preprocessing process. In this paper, we propose an efficient algorithm to remove noise in high density of Salt and Pepper noise. The proposed algorithm removes noise by gradually expanding the filtering mask according to the density of the noise, and shows excellent noise cancellation performance even in a high density region. In order to evaluate the performance of the proposed algorithm, we compared and analyzed the existing method and the proposed algorithm through simulation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.165-168
• /
• 2021

Recently, the spread of deep learning environments has increased the importance of dataset generation. In this paper, we aim to design and implement a method for capturing rotating images of objects and performing object tracking on them for efficient dataset generation. The method implemented in this paper is to obtain image data by rotating objects to capture multiple angles of objects, detect and track objects through background removal and difference image processing techniques, showing them on screen to monitor object tracking results in the current frame. It was then implemented to return object location data within the image for use as a dataset.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.25-28
• /
• 2024

본 논문에서는 LiDAR 데이터와 스테레오 이미지를 융합하여 고품질 3D 표현을 생성하는 새로운 접근 방식을 제안한다. 제안하는 방법은 LiDAR 포인트 클라우드와 스테레오 비전을 통해 3D 포인트를 추출하는 것에서 시작하며, 이후 스테레오 비전 포인트 클라우드의 스케일을 LiDAR 스케일에 맞추는 조정 과정을 거친다. 스케일이 조정된 스테레오 포인트와 LiDAR 데이터를 초기 융합하여 두 가지 모달리티의 장점을 모두 활용한 포괄적인 포인트 클라우드를 생성한다. 융합된 포인트 클라우드를 정제하기 위해, DBSCAN과 같은 클러스터링을 통한 노이즈 제거와 포인트 그룹화, 그리고 LiDAR 데이터를 기준으로 스테레오에서 추출한 포인트들을 정밀하게 맞추기 위한 회귀 모델을 결합한 하이브리드 기법을 도입한다. 정제된 포인트 클라우드는 3D Gaussian Splatting 초기화를 위한 기초로 사용되며, 각 포인트를 초기 가우시안 값으로 설정하고 다양한 뷰포인트에서의 렌더링 결과를 바탕으로 가우시안 파라미터를 최적화한다. 최적화된 3D 가우시안을 활용하여 다양한 시점에서 장면을 렌더링하고, 이를 통해 연속적이고 풍부한 3D 장면 표현을 생성한다. 본 연구는 일반적인 새로운 뷰 합성(general novel view synthesis) 문제에 대한 중요한 개선을 달성하여, 컴퓨터 비전, 자율주행, 가상현실과 같은 분야에서의 응용 가능성을 보여준다.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.3
• /
• pp.151-162
• /
• 2020

This paper presents an acoustic inversion method for estimating the bubble size distribution. The estimation error of the attenuation coefficient represented by a Fredholm integral equation of the first kind is defined as an objective function, and an optimal solution is found by applying the Levenberg-Marquardt (LM) method. In order to validate the effectiveness of the inversion method, numerical simulations using two types of bubble distribution are performed. In addition, a series of experiments are carried out in a water tank (1.0 m × 0.54 m × 0.6 m), using bubbles generated by three different generators. Images of the distributed bubbles are obtained by a high-speed camera, and the insertion losses of the bubble layer are measured using a source and a hydrophone. The image is post-processed to glance a distribution characteristics of each bubble generator. Finally, the size distribution of bubbles is estimated by applying the inversion method to the measured insertion loss. From the inversion results, it was observed that the number of bubbles increases exponentially as the bubble size decreases, and then increases again after the local peak at 70 ㎛ - 120 ㎛.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.7
• /
• pp.3454-3459
• /
• 2013

Sumi-e is one of the art work that uses not only ink line but also color painting. This technique is well known as a representative Asian painting style and widely used in movie, advertisement poster and various effect in camera device. In this paper, we propose an algorithm that can generate result image with Sumi-e effects of example image based on computer graphics and image processing techniques. For this, we pass two steps. The first is painting expression step. We used texture transfer technique to generate result with texture effect of reference image by analyzing numerically. The second step is ink-painting effect generation step. We express ink-painting effect in outline by considering intensity variation in edge of example image. Our algorithm can express various Sumi-e style based on selected reference image. So it can be utilized to various contents generating research.

• Journal of Internet Computing and Services
• /
• v.4 no.6
• /
• pp.57-67
• /
• 2003

The recent rapid development of multimedia and optical technologies brings great attention to application systems to process facial Image features. The previous research efforts in facial image processing have been mainly focused on the recognition of human face and facial expression analysis, using front face images. Not much research has been carried out Into image-based detection of face direction. Moreover, the existing approaches to detect face direction, which normally use the sequential Images captured by a single camera, have limitations that the frontal image must be given first before any other images. In this paper, we propose a method to detect face direction by using facial features such as facial trapezoid which is defined by two eyes and the lower lip. Specifically, the proposed method forms a facial direction formula, which is defined with statistical data about the ratio of the right and left area in the facial trapezoid, to identify whether the face is directed toward the right or the left. The proposed method can be effectively used for automatic photo arrangement systems that will often need to set the different left or right margin of a photo according to the face direction of a person in the photo.

• The Journal of the Korea Contents Association
• /
• v.9 no.5
• /
• pp.13-23
• /
• 2009

One of the main purpose of multiple-view image processing technology is support realistic 3D image to device user by using multiple viewpoint display devices and compressed data restoration devices. This paper proposes a multiple view image synthesis scheme based on RAM disk which makes possible to browse 3D images generated by applying effective composing method to real time input stereo images. The proposed scheme first converts input images to binary image. We applies edge detection algorithm such as Sobel algorithm and Prewiit algorithm to find edges used to evaluate disparities from images of 4 multi-cameras. In addition, we make use of time interval between hardware trigger and software trigger to solve the synchronization problem which has stated ambiguously in related studies. We use a unique identifier on each snapshot of images for distributed environment. With respect of performance results, the proposed scheme takes 0.67 sec in each binary array. to transfer entire images which contains left and right side with disparity information for high quality 3D image browsing. We conclude that the proposed scheme is suitable for real time 3D applications.