• 한국멀티미디어학회논문지
• /
• 제20권2호
• /
• pp.163-169
• /
• 2017

There are various algorithms evaluating a threshold for image segmentation. Among them, Otsu's algorithm sets a threshold based on the histogram. It finds the between-class variance for all over gray levels and then sets the largest one as Otsu's optimal threshold, so we can see that Otsu's algorithm requires a lot of the computation. In this paper, we improved the amount of computational needs by using estimated Otsu's threshold rather than computing for all the threshold candidates. The proposed algorithm is compared with the original one in computation amount and accuracy. we confirm that the proposed algorithm is about 29 times faster than conventional method on single processor and about 4 times faster than on parallel processing architecture machine.

• 한국지능시스템학회논문지
• /
• 제20권3호
• /
• pp.368-374
• /
• 2010

영상 분할 (Image Segmentation)은 패턴 인식, 환경 인식, 문서 분석을 위한 영상 처리 과정에서 가장 기본적인 단계이다. 영상 분할 방법들 중 Otsu의 영상의 정규화된 히스토그램의 분포 정보를 이용하여 클래스 간의 분산을 최대화 시키는 임계치값을 결정하는 자동 임계치값 선정방법이 가장 잘 알려진 방법이다. Otsu의 방법은 영상의 전 영역에 대한 히스토그램을 분석함으로써 영상의 부분적인 특성을 반영하여 임계치값을 결정하기는 어렵다. 본 논문에서는 이 어려움 해소하기 위하여 Context Fuzzy c-Means 알고리즘을 이용하여 영상을 여러 개의 부분 영역으로 나누고, 정의된 부 영역에 영상 분할 기법을 적용함으로써 부 영역들에 적합한 여러 개의 임계치값을 계산함으로써 영상 분할 성능을 개선하고자 하였다.

• 동력기계공학회지
• /
• 제20권4호
• /
• pp.32-37
• /
• 2016

This paper proposes two measurement methods for injured rate of fish surface using color image segmentation method based on K-means clustering algorithm and Otsu's threshold algorithm. To do this task, the following steps are done. Firstly, an RGB color image of the fish is obtained by the CCD color camera and then converted from RGB to HSI. Secondly, the S channel is extracted from HSI color space. Thirdly, by applying the K-means clustering algorithm to the HSI color space and applying the Otsu's threshold algorithm to the S channel of HSI color space, the binary images are obtained. Fourthly, morphological processes such as dilation and erosion, etc. are applied to the binary image. Fifthly, to count the number of pixels, the connected-component labeling is adopted and the defined injured rate is gotten by calculating the pixels on the labeled images. Finally, to compare the performances of the proposed two measurement methods based on the K-means clustering algorithm and the Otsu's threshold algorithm, the edge detection of the final binary image after morphological processing is done and matched with the gray image of the original RGB image obtained by CCD camera. The results show that the detected edge of injured part by the K-means clustering algorithm is more close to real injured edge than that by the Otsu' threshold algorithm.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2012년도 추계학술대회
• /
• pp.259-262
• /
• 2012

In this paper, we propose a new fast algorithm for effective scene change detection. The proposed algorithm exploits Otsu threshold matching technique, which was proposed earlier. In this method, the current and the reference frames are divided into square blocks of particular size. After doing so, the pixel histogram of each block is generated. According to Otsu method, every histogram distribution is assumed to be bimodal, i.e. pixel distribution can be divided into two groups, based on within-group variance value. The pixel value that minimizes the within-group variance is said to be Otsu threshold. After Otsu threshold is found, the same procedure is performed at the reference frame. If the difference between Otsu threshold of a block in the current frame and co-located block in the reference frame is larger than predefined threshold, then a scene change between those two blocks is detected.

• 한국멀티미디어학회논문지
• /
• 제13권7호
• /
• pp.1062-1075
• /
• 2010

의료 영상 분야에서 영상의 분할 및 특성의 추출을 위하여 명암도 차이를 이용하는 방법이 널리 사용되고 있으며, 임계값을 결정한 뒤 이를 기준으로 영상을 이진화하는 임계값 방식이 잘 알려져 있다. 임계값 방식 중 자주 사용되는 방식이 임계값을 선택하는 데 효율적이면서, 효과적인 선정 기준을 제시하고 있는 Otsu 알고리즘이다. 하지만 흉부 X-ray 영상에 대해서는 Otsu 알고리즘의 적용으로 좋은 영상 분할 결과를 얻을 수 없다. 이는 폐 영역 주변에는 갈비뼈나 혈관과 같은 다양한 기관이 존재하여 따라서 명암도 레벨의 분포가 불명확하기 때문이다. 이러한 불명료성을 개선하기 위하여, 본 논문에서는 X-ray 영상의 배경을 배제한 후 Otsu 알고리즘을 적용하고, 명암 레벨 지도를 생성한 후, 이를 이용하여 X-ray 영상을 분할하는 효과적인 폐 영역 추출 알고리즘을 제시한다. 제안한 방법의 효과를 검증하기 위해 제안한 방법과 기존의 1차원 및 2차원 Otsu 알고리즘, 그리고 전문가의 육안 분할 결과와 비교하였다. 실험 결과, 제안한 방법이 기존 Otsu 방법에 비해 더 정확하게 폐 영역을 추출하였으며, 육안 분할 결과와 거의 비슷한 결과를 보여 주었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.287-287
• /
• 2000

As a method to control the bacteria number in adequate level, a real time control system based on microscope image processing measurement for the bacteria is adopted. For the experiment, Ammonia-oxidizing bacteria such as Acinetobacter sp. are used. This paper proposed hybrid method combined watershed algorithm with adaptive automatic thresholding method to enhance segmentation efficiency of overlapped image. Experiments was done to show the effectiveness of the proposed method compared to traditional Otsu's method, Otsu's method with adaptive automatic thresholding method and human visual method.

• International Journal of Automotive Technology
• /
• 제7권2호
• /
• pp.227-235
• /
• 2006

In this paper a robust image processing method with modified Otsu algorithm to recognize the road lane for a real-time controlled autonomous vehicle is presented. The main objective of a proposed method is to drive an autonomous vehicle safely irrespective of road image qualities. For the steering of real-time controlled autonomous vehicle, a detection area is predefined by lane segment, with previously obtained frame data, and the edges are detected on the basis of a lane width. For stable as well as psudo-robust autonomous driving with "good", "shady" or even "bad" road profiles, the variable threshold with modified Otsu algorithm in the image histogram, is utilized to obtain a binary image from each frame. Also Hough transform is utilized to extract the lane segment. Whether the image is "good", "shady" or "bad", always robust and reliable edges are obtained from the algorithms applied in this paper in a real-time basis. For verifying the adaptability of the proposed algorithm, a miniature vehicle with a camera is constructed and tested with various road conditions. Also, various highway road images are analyzed with proposed algorithm to prove its usefulness.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2009년도 정보 및 제어 심포지움 논문집
• /
• pp.40-42
• /
• 2009

This paper presents an algorithm, which is based on Otsu's method, for accurate and robust endpoint detection for speech recognition under noisy environments. The features are extracted in time domain, and then an optimal threshold is selected by minimizing the discriminant criterion, so as to maximize the separability of the speech part and environment part. The simulation results show that the method play a good performance in detection accuracy.

• 전기전자학회논문지
• /
• 제24권1호
• /
• pp.127-131
• /
• 2020

클러스터링을 이용한 대표적인 영상 분할 방법으로 Fuzzy C-Means(FCM) 알고리즘을 많이 사용하는데, FCM은 영상의 공간을 픽셀 값이 비슷한 클러스터 영역으로 분할하므로 분할 시간이 많이 소요된다. 특히 웹이 보편화된 현재 사용자들의 다양한 패턴을 분석하기 위한 처리 속도 문제는 더욱 중요하다. 이러한 속도 문제를 해결하기 위해 본 논문에서는 Otsu의 영상 히스토그램의 임계값과 FCM으로 영상을 분할하는 개선된 FCM(Improved FCM : IFCM) 알고리즘을 제안한다. 제안방법은 Otsu의 클래스 간의 분산을 최대화 시키는 임계값을 결정하여 FCM에 적용하고 영상을 분할하였다. IFCM은 기존의 FCM에 비해 영상 분할 시간을 단축시켜 성능이 향상되었음을 실험을 통해 보인다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 춘계학술대회논문집B
• /
• pp.670-675
• /
• 2001

This paper is to propose image segmentation method based on chamfer algorithm. First, we get original image from CCD camera and transform it into gray image. Second, we extract maximum gray value of background and reconstruct and eliminate the background using surface fitting method and bilinear interpolation. Third, we subtract the reconstructed background from gray image to remove noises in gray image. Fourth, we transform the subtracted image into binary image using Otsu's optimal thresholding method. Fifth, we use morphological filters such as areaopen, opening, filling filter etc. to remove noises and isolated points. Sixth, we use chamfer distance or Euclidean distance to this filtered image. Finally, we use watershed algorithm and count microorganisms in image by labeling. To prove the effectiveness, we apply the proposed algorithm to one of Ammonia-oxidizing bacteria, Acinetobacter sp. It is shown that both Euclidean algorithm and chamfer algorithm show over-segmentation. But Chamfer algorithm shows less over-segmentation than Euclidean algorithm.