• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.6
• /
• pp.1160-1169
• /
• 1997

This paper presents an algorithm to separate vowels from consonants in Korean characters captured in noisy environment andto recognize them. The algorithm has been originally developed for recognition of the usage code (which is represented by a single Korean character) in the license plates of Korean vehicles. It, however, could be easily adopted to other applications with minor changes, in which character recognition is needed and the environment is noisy. The key ideas of the algorithm are to localize the vowels utilizing Hough transformation and to separate the vowels from consonants utilizing mathematical morphology. We observed that the presented algorithm effectively separates vowels even if the vowels and consonants are joined together after thresholding. We also observed that our algorithm outperforms some conventional algorithms especially when the input images are noisy. The details of the comparison study are presented in the paper.

• Journal of Korea Multimedia Society
• /
• v.18 no.10
• /
• pp.1149-1156
• /
• 2015

This paper proposes the hand contour detector which is robust to noises. Existing methods reduce noises by applying morphology to extracted edges, detect finger tips by using the center of hands, or exploit the intersection of curves from hand area candidates based on J-value segmentation(JSEG). However, these approaches are so vulnerable to noises that are prone to detect non-hand parts. We propose the noise tolerant hand contour detection method in which non-skin area noises are removed by applying skin area detection, contour detection, and a threshold value. By using the implemented system, we observed that the system was successfully able to detect hand contours.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2387-2390
• /
• 2003

움직이는 물체의 자동 분할은 컴퓨터 비젼의 여러 응용분야에서 중요한 문제로 대두되고 있다. 본 논문에서는 감시 시스템에서 에지와 적응적 임계값을 이용한 효과적인 자동 움직임 분할 방법을 제안하였다. 먼저 연속 영상에서 현재 영상과 배경 영상과의 차를 얻어서 그 히스토그램을 만든다. 이 때 앞에서 얻은 히스토그램은 영상 잡음의 평균이 0 인 가우시안 분포를 가진다고 가정한다. 그리고, 이 히스토그램을 이용하여 영상잡음의 분산을 찾는다 이 분산 값을 이용하여 적응적 임계값과 움직임 영역창을 결정한다. 적응적 임계값에 의한 결과 영상에서 움직이는 물체를 분할하기 위해 본 논문에서는 움직임 영역창을 이용하는 방법을 제안하였다. 이 움직임 영역창에 의해 더욱 효과적인 움직임 분할이 이루어진다. 또, 잡음의 제거를 위해 수학적 모폴로지(mathematical morphology)와 화소의 연결성이 이용된다.

• International Journal of Aeronautical and Space Sciences
• /
• v.9 no.2
• /
• pp.114-120
• /
• 2008

In the vision-based recovery phase, a terminal guidance for the blended-wing UAV requires visual information of high accuracy. This paper presents the light source target design and detection algorithm for vision-based UAV recovery. We propose a recovery target design with red and green LEDs. This frame provides the relative position between the target and the UAV. The target detection algorithm includes HSV-based segmentation, morphology, and blob processing. These techniques are employed to give efficient detection results in day and night net recovery operations. The performance of the proposed target design and detection algorithm are evaluated through ground-based experiments.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.802-804
• /
• 2005

제안된 연구에서는 기존의 일반 CT(Computerized tomography) 영상이 아닌 MDCT(Multi Detector CT) 영상을 이용하여 장기 추출에 관한 연구를 진행하였다. 조영제를 이용한 복부 MDCT 영상으로부터 모폴로지(morphology) 기법을 통해 간에 근접한 노이즈를 제거하고, 기존의 Otsu threshold를 개선하여 간의 명암값 분포를 구분할 수 있는 임계치를 구하였다. 찾아진 임계치를 이용하여 영상을 이진화하고, 최종적으로 위치정보를 이용하여 간에 해당하는 부분들을 추출하였다. 이러한 방식은 명암값과 위치정보를 이용하여 간을 추출한 후 다시 노이즈 문제를 해결하는 기존의 알고리즘과 비교했을 때, 처리 방식이 단순해지고 속도가 향상되었다. 추출된 간은 간 이식술이나 절제술에 필요한 간 내부의 혈관 인식과 간의 부분체적 계산 연구에 중요한 정보로 사용될 수 있을 것이다.

• Proceedings of the KOSOMBE Conference
• /
• v.1993 no.11
• /
• pp.17-19
• /
• 1993

정자의 형태 분석은 운동 분석과 더불어 불임의 원인 규명 및 진단에 중요한 정보를 제공한다. 본 연구에서는 Diff-Quick 염색된 정자의 영상에 대해, 제안된 Image segmentation 방법을 적용해 정자 형태 특성을 검출해 내는 알고리즘을 구현했다. 비디오 신호로 보내진 정자 영상을 디지탈화하고, 정확한 테두리를 찾은 뒤, 정자 머리 부분의 형태 정보를 추출한 후, 그 특성을 나타내는 parameter 들을 추정하였다.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.923-926
• /
• 2003

In this paper we propose a scheme to generate DTM and detect buildings on DSM generated from LIDAR data. Two stages are performed. The first stage is to perform object segmentation by using two morphology operations namely, flattening and H-Dome transformation. After filtering out the object points above the ground, we used the non-object points to generate DTM. The second stage is to detect buildings from the objects by analyzing differential slopes. The test data is in raster form with 1m spacing around Hsin-Chu Scientific Area in Taiwan. The mean error is -0.16m and the RMSE is 0.45m for DTM generation. The successful rate for building detection is 87.7%.

• Journal of Drive and Control
• /
• v.19 no.3
• /
• pp.16-22
• /
• 2022

Crushers are equipment that crush natural stones, to produce aggregates used at construction sites. As the crusher proceeds, the inner liner becomes worn, causing the size of the aggregate produced to gradually increase. The vision sensor-based aggregate analysis system analyzes the size and distribution of aggregates in production, in real time through image analysis. This study developed an algorithm that can segmentate aggregates in images in real time. using image preprocessing technology combining various filters and morphology techniques, and aggregate region characteristics such as convex hull and concave hull. We applied the developed algorithm to fine aggregate, intermediate aggregate, and thick aggregate images to verify their performance.

• Journal of Biomedical Engineering Research
• /
• v.19 no.6
• /
• pp.571-575
• /
• 1998

One of the most significant features of diagnostic ultrasonic instruments is to provide real time information of the soft tissues movements. Echocardiogram has been widely used for diagnosis of heart diseases since it is able to show real time images of heart valves and walls. However, the currently used ultrasonic images are deteriorated due to presence of speckle noises and image dropout. Therefore, it is very important to develop a new technique which can enhance ultrasonic images. In this study, a technique which extracts enhanced binary images in echocardiograms was proposed. For this purpose, a digital moving image file was made from analog echocardiogram, then it was stored as 8-bit gray-level for each frame. For an efficient image processing, the region containing the heat septum and tricuspid valve was selected as the region of interest(ROI). Image enhancement filters and morphology filters were used to reduce speckle noises in the images. The proposed procedure in this paper resulted in binary images with enhanced contour compared to those form the conventional threshold technique and original image processing technique which can be further implemented for the quantitative analysis of the left ventricular wall motion in echocardiogram by easy detection of the heart wall contours.

• Journal of the Korea Society for Simulation
• /
• v.27 no.2
• /
• pp.1-10
• /
• 2018

This paper presents an autonomous detection and aiming point computation of a battle tank by using RGB images. Maximally stable extremal regions algorithm was implemented to find features of the tank, which are matched with images extracted from streaming video to figure out the region of interest where the tank is present. The median filter was applied to remove noises in the region of interest and decrease camouflage effects of the tank. For the tank segmentation, k-mean clustering was used to autonomously distinguish the tank from its background. Also, both erosion and dilation algorithms of morphology techniques were applied to extract the tank shape without noises and generate the binary image with 1 for the tank and 0 for the background. After that, Sobel's edge detection was used to measure the outline of the tank by which the aiming point at the center of the tank was calculated. For performance measurement, accuracy, precision, recall, and F-measure were analyzed by confusion matrix, resulting in 91.6%, 90.4%, 85.8%, and 88.1%, respectively.