Earticle

This paper presents two methods for buildings extraction in Very High Resolution (VHR) remotely sensed multispectral (4 band) images based on supervised and unsupervised segmentation using different image properties. The proposed approach for unsupervised or automatic building detection involves four stages, primarily, filtering to smoothen and enhance the objects present and sharpen the details. Secondly, a binary mask creation over which edge detection is applied. Edge linking is done to preserve information about the object. Lastly region properties like area, perimeter, etc are applied on the prepared mask and buildings are detected. The semi-automated or supervised method uses advanced color based segmentation algorithm to extract the buildings tops. This technique creates a number of masks based on segmentation and uses region properties based on color. Experiments are made on VHR images captured from satellites of commercial companies like Digital Globe and Geo Eye. Results of both methods are compared with respect to various accuracy measures at the end. The results illustrate that supervised algorithm using color property produces more accurate object delineation.

Using color linear array CCD (Charge Coupled Device) to collect the information of image has an important meaning in the research of the machine vision processing. It now becomes the focus of study on CCD image acquisition using FPGA (Field Programmable Gate Array) to implement color linear array CCD-driven design. The image acquisition and processing of the rice color sorter as the research background in this paper, by using FPGA to achieve the color linear array CCD driven control, as well as simulated with Quartus II 8.0. In order to be able to extract the color linear array CCD output useful information of the color images, through study the RGB model and the HSI (Huge, Saturation, Intensity) model of color space model, the HSI space mode is selected as the color image processing space model, and design the suitable algorithm-FCM (Fuzzy C-Means) clustering algorithm in HSI space for color image segmentation, through a fuzzy Toolbox in MATLAB to accomplish the FCM algorithm, at the same time, through the simulation realized the FCM algorithm. The simulation shows that the algorithm can segment the image useful information well.

Skin color segmentation plays an important role in the biological feature recognition system. In order to improve the system adaptability to the complex environment, a new model of Cr*a*b by combing the YCbCr with Lab model is proposed. Firstly, establish the three-dimensional coordinate system based on the Cr, a, b axes, the two-two coordinates form the directional vectors by using the origin as the starting point. Then calculate the angles between the three vectors. Finally, realize the purpose of skin color segmentation by reducing the dimensions of angle data. Experimental results show that this algorithm can improve the over sensitive of YCbCr model to red and yellow, the defect of the skin color weak clustering in the Lab model is also overcame. Even if for the image with complex background environment, the percent of average accurate detection is about 87.7%. The proposed algorithm can get the high skin color segmentation rate and has good practicability.

Color image segmentation algorithms are proposed based on granular computing clustering (GrCC). Firstly, the atomic hyperspherical granule is represented as the vector including the RGB value of pixel of color image and radii 0. Secondly, the union operator of two hyperspherical granules is designed to obtain the larger hyperspherical granule compared with these two hyperspherical granules. Thirdly, the granular computing clustering is developed by the union operator and the user-defined granularity threshold . Global Consistency Error (GCE), Variation of Information (VI), Rand Index (RI), and Loss Entropy (ΔEn) are used to evaluate the segmentations. Segmentations of the color images selected from internet and BSD300 show that segmentations by GrCC speed up the segmentation process and achieve the better segmentation performance compared with Kmeans and FCM segmentations.

Aiming at the problems that current skin-color detection segmentation technologies have unsatisfied segmentation results under conditions of complex illumination or backgrounds, we present a new method based on YCbCr color space and K-means clustering algorithm for segmentation hand gesture. Firstly, image in RGB color space is converted to YCbCr color space; and then YCbCr color space of image is divided into luminance Y and chrominance Cb and Cr. Lastly, the binary image is achieved by clustering values of chrominance using k-means clustering algorithm, and hand gesture segmentation is completed by conducting morphological process of binary image obtained. The experimental results illustrate that the proposed method can segment hand gestures from complex backgrounds and obtain segmentation results. The phenomena of similar skin color interference and skin color overlapping are solved with this method effectively. In addition, it is robust to illumination condition.

In order to improve the problems of segmentation accuracy and real-time existing in dynamic hand gesture under complex backgrounds, this paper presents a kind of dynamic hand gesture segmentation method based on improved Kalman filter and weighted skin-color model. Firstly, improved Kalman filter is utilized to process hand gesture image of hand gesture video sequences and get rough hand gesture results. Secondly, weighted skin-color model is applied to process rough results of hand gesture segmentation and segment hand gesture. Finally, morphological method is utilized to deal with gesture segmentation result, getting rid of the holes in the hand gesture’s binary image to realize the segmentation of dynamic hand gesture. Experiments show that the proposed method can segment hand gesture from dynamic hand gesture video sequences with complex backgrounds effectively. And the accuracy of hand gesture segmentation is high.

The accurate segmentation and extraction of bamboo cross-section image has a vital role on bamboo processing automation. The Lab color space in accordance with the color features of the bamboo wood cross-section is chosen in this paper. The bamboo cross-section image segmentation and extraction algorithm based on the clustering theory is proposed. The algorithm firstly takes advantage of the character that the colors represented by channel a and channel b of the Lab color space accord with the color of the bamboo wood, to be combined feature vector. Then the algorithm uses the k-mean clustering algorithm to classify the eigenvectors to realize the segmentation of the bamboo wood cross-section. At last the circle fitting algorithm is used to realize the final frame of the bamboo wood cross-section. The results of the experiments show that the algorithm can be used to realize the complete segmentation of the cross-section image of bamboo wood, and to frame the results correctly, the time performance of which can meet the requirements of the subsequent processing.

In this study, we propose the new color-based image retrieval technique using the representative colors of images and their ratios to a total image size obtained through color segmentation in HSV color space. Color information of an image is described by reconstructing the color histogram of an image through Gaussian modelling to its representative colors and ratios. And the similarity between two images is measured by histogram intersection. The proposed method is compared with the existing methods by performing retrieval experiments for various 1280 trademark image database.

본 논문에서는 칼라 영역으로 확장된 블록단위 영상분할 알고리듬을 제안한다. 즉, 기존의 휘도 성분을 기반으로 한 블록단위 알고리듬을 HSV 칼라 성분을 기반으로 하는 칼라 영상분할로 확장한다. 기존의 수학적 형태학(mathematical morphology)에 기반한 영상분할기법이나 블록단위 영상분할 기법들이 밝기 정보만을 활용했던 것에 대해 제안된 블록단위 영상분할 기법은 밝기뿐만이 아니라 칼라 성분도 고려하여 영상분할의 정확도를 향상시켰다. 실험의 결과 휘도 성분만을 고려한 영상분할 결과 보다 칼라 성분을 사용한 영상분할의 결과가 더 정확함을 알 수 있었다.

본 논문에서는 수학적 모폴로지 연산과 가우시안 혼합 모형에 기초한 새로운 칼라 영상 분할 알고리즘을 제안한다. 우리는 혼합 모형에서 구성 성분의 수를 결정하고, 각 구성 성분의 중심값을 계산하는데 모폴로지의 연산과 라벨링 연산을 이용한다. 그리고 칼라 특징 벡터의 확률 모형으로 가우시안 혼합 모형을 사용하고, 이들의 모수 값들을 추정하는데 결정적 어닐링 EM알고리즘을 사용한다. 최종적으로 혼합 모형으로부터 계산된 사후 확률을 이용하여 칼라 영상을 분할한다. 실험 결과를 통하여 모폴로지 연산이 혼합모형의 수를 자동으로 결정하고 각 성분의 모드를 계산하는데 아주 효율적인 방법임을 보였고, 또한 결정적 어닐링 EM 알고리즘에 의하여 추정된 가우시안 혼합 모형을 사용하여 계산된 사후 확률에 의한 영상 분할 방법이 기존의 분할 알고리즘보다 정확한 분할 방법임을 보였다.

In this paper, we present a new segmentation algorithm for color images based on mathematical morphology and a Gaussian mixture model(GMM). We use the morphological operations to determine the number of components in a mixture model and to detect their modes of each mixture component. Next, we have adopted the GMM to represent the probability distribution of color feature vectors and used the deterministic annealing expectation maximization (DAEM) algorithm to estimate the parameters of the GMM that represents the multi-colored objects statistically. Finally, we segment the color image by using posterior probability of each pixel computed from the GMM. The experimental results show that the morphological operation is efficient to determine a number of components and initial modes of each component in the mixture model. And also it shows that the proposed DAEM provides a global optimal solution for the parameter estimation in the mixture model and the natural color images are segmented efficiently by using the GMM with parameters estimated by morphological operations and the DAEM algorithm.

본 논문에서는 DCT 데이터에서 영상 데이터로의 해독 및 이진화 과정을 생략하고 컬러 영상의 DCT 관련 원자료를 사용하는 방법에 기반을 둔 매우 빠르고 안정적인 문자열 구간 분리 모형을 제안하였다. DCT 블록에 저장된 DC 및 3개의 주요 AC 변수들을 조합하여 축소된 저해상도 회색 영상을 만들고 횡렬 및 종렬 투영법을 통해 얻어진 픽셀 값의 히스토그램을 분석하여 문자 열 구간 사이에 존재하는 백색의 띠 공간을 찾아내었다. 이 과정 중 탐색되지 않은 문자 열 구간은 마코프 모델을 사용하여 숨겨진 주기를 찾아내어 복원하였다. 본 논문에 실험 결과를 제시하였으며 기존의 방법보다 약 40 - 100배 빠른 방법임을 입증하였다.

We presented a very fast and robust method of text line segmentation based on the DCT blocks of color image without decompression and binary transformation processes. Using DC and another three primary AC coefficients from block DCT we created a gray-scale image having reduced size by 8x8. In order to detect and locate white strips between text lines we analyzed horizontal and vertical projection profiles of the image and we applied a direct markov model to recover the missing white strips by estimating hidden periodicity. We presented performance results. The results showed that our method was 40 - 100 times faster than traditional method.

많은 연구 데모용 프로그램들과 상업적 응용물들이 얼굴 검출과 얼굴 인식 시스템들을 개발하기 위해 시도되고 있다. 인간의 얼굴 검출은 접근 제어 및 비디오 감시 시스템, 휴먼 컴퓨터 인터페이스, 신원 인증 등과 같은 많은 응용 프로그램들에 중요한 역할을 한다. 일반적으로 스킨 영역 분할 후 배경과 연결된 얼굴, 스킨 칼라로 인한 연결된 얼굴들, 여러 개의 작은 부분들로 분할된 하나의 얼굴과 같은 몇 가지 특별한 문제점들이 있다. 많은 얼굴 검출 기법들이 첫 번째 와 두 번째 문제를 해결하도록 허락되어진다. 그러나 세 번째 문제에서 다른 조명 효과들로 인해서 여러 영역들로 분할된 하나의 얼굴이 검출되어지는 것은 쉽지가 않다. 그러므로 우리는 기존 영역 분할 알고리즘은 이용될 수 없기 때문에 이 문제를 해결하기 위해 효율적인 수정된 스킨 분할 알고리즘을 제안한다. 본 알고리즘은 전체 영상에 대해 피부 영역을 검출한 후 피부 분할 알고리즘을 사용하여 얼굴 후보 영역들을 생성한다. 각 얼굴 피부 후보 영역에 대해 그림자 등의 조명 효과로 인해 한 명의 얼굴이 여러 영역으로 분할되는 경우를 처리하기 위해 동차적 영역간의 인접성을 활용하여 하나의 큰 영역으로 만드는 병합 작업을 시도하였다. 다른 크기의 얼굴 검출을 위해 다양한 가변 크기의 탐색 윈도우와 선택된 각 얼굴 후보 영역에 얼굴이 존재하는지를 판단하기 위해 역전파 알고리즘에 기반한 얼굴 검출 분류기를 사용하였다.

Many research demonstrations and commercial applications have been tried to develop face detection and recognition systems. Human face detection plays an important role in applications such as access control and video surveillance, human computer interface, identity authentication, etc. There are some special problems such as a face connected with background, faces connected via the skin color, and a face divided into several small parts after skin region segmentation in generally. It can be allowed many face detection techniques to solve the first and second problems. However, it is not easy to detect a face divided into several parts of regions for reason of different illumination conditions in the third problem. Therefore, we propose an efficient modified skin segmentation algorithm to solve this problem because the typical region segmentation algorithm can not be used to. Our algorithm detects skin regions over the entire image, and then generates face candidate regions using our skin segmentation algorithm For each face candidate, we implement the procedure of region merging for divided regions in order to make a region using adjacency between homogeneous regions. We utilize various different searching window sizes to detect different size faces and a face detection classifier based on a back-propagation algorithm in order to verify whether the searching window contains a face or not.

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.

This paper proposes a digital algorithm for gray-scale/color image segmentation of real-time video signals and a cell-network-based implementation architecture in state-of-the-art CMOS technology. Through extrapolation of design and simulation results we predict that about 300$\times$300 pixels can be integrated on a chip at 100nm CMOS technology, realizing very high-speed segmentation at about 1600sec per color image. Consequently real-time color-video segmentation will become possible in near future.

멀티미디어 정보표현 기술인 MPEG-7 표준이 빠른 속도의 진전을 보임에 따라 이를 활용한 검색 기술 개발도 활발히 진행 중에 있다 방대한 량의 동영상 내용 검색 기술 연구에 있어서 우선적으로 고려되어야 할 부분이 내용이 연속되는 프레임들의 분류이다. 이를 위해서는 물리적인 장면전환이 이루어지는 부분에 대한 실시간 자동 cut detection 기술 및 이 컷 프레임 영상에 대한 내용 기술을 자동적으로 수행할 필요성이 있다. 각 컷 프레임의 자동 내용 기술의 전처리로써 본 논문에서는 장면전환이 생기는 프레임의 영상의 어떠한 정보도 사전 정보로 취하지 않고 사용자의 개입이 없는 상황에서 영상의 컬러 특성 및 에지 정보만을 가지고 shape 영역 segmentation을 자동으로 실행하는 방법을 제안한다. 제안한 방법의 성능은 segmentation된 영상과 원 영상과의 영역비교를 통한 유사도에 의해 평가하며, 시뮬레이션 결과에서 제안한 알고리즘은 평균 90%이상의 영역 분할이 정확하게 됨을 알 수 있었고, 컬러의 구분이 명확하지 않은 자연영상에서도 robust한 segmentation 결과를 나타냄을 본 연구를 통하여 알 수 있었다.

This paper describes a new semi-automatic segmentation algorithm based on color information. Semi-automatic segmentation mainly consists of intra-frame segmentation and inter-frame segmentation. While intra-frame segmentation extracts video objects of interest from boundary information provided by the user and intensity information of the image, inter-frame segmentation partitions the image into the video objects and background by tracking the motion of video objects. For inter-frame segmentation, color information (Y, Cb and Cr) of the current frame can be used efficiently in order to find the exact boundary of the video objects. In this paper we propose a new region growing algorithm which can maximize the ability of region differentiation, while preserving features of each color component.

In this paper, video image segmentation algorithm based on color histogram and change detector is proposed. Color histograms are calculated from both changed region which is detected in the previous and current frame and unchanged region. With each histogram, modes and valleys are detected. Then, color vectors are calculated by averaging pixels in modes. Markers are extracted by labeling color vectors that represent modes, the watershed algorithm is applied to determine uncertain region. In growing region, the root mean square(RMS) of the distance between average pixel in marker region and adjacent pixel is used as a measure. The proposed algorithm based on color histogram and change detector segments video image fastly and effectively. And simulation results show that the proposed method determines the exact boundary between background and foreground.

내용을 기반으로 하는 영상검색에 있어 색상과 물체의 특징은 중요한 요소로서, 지금까지의 검색 기법들은 이들을 중심으로 연구가 진행되어 왔으며, 이들을 추출하기 위해서는 color 영상에서의 배경과 물체의 분리는 선행되어야 할 중요한 과제이다. color 영상에서 물체를 분리 하고자 하는 여러 가지 시도가 있었으나, 대부분 clustering 에 준하고 있으며, 처리시간이나 결과에 있어서 그다지 좋은 효과를 내지 못하는 것도 사실이다. 따라서, 영상검색을 위한 물체의 분리 기법으로서는 적합하지 않다. 본 논문에서는 물체가 영상의 중심에 주로 위치한다는 점에 착안한 방법을 응용하여 영상의 외곽에 존재하는 색상뿐만 아니라 명암까지 분석하여, 배경을 구성하는 화소들의 색상 및 명암과 동일하지 않은 색상들로 이루어진 부분을 물체로 판단, 추출하는 기법에 대해 설명하고, edge를 추출해낸 영상의 정보와 합성하여 최적의 물체를 찾아 검색을 하는 기법에 대하여 기술하였다.

본 논문에서는 스테레오 매칭 시 발생하는 신뢰도가 낮은 부분을 컬러와 명도의 다중 임계값에 기반한 영상 분할 기법을 통해 보정하는 방법을 제안한다. 스테레오 매칭은 좌측 영상 위의 한 점과 대응하는 점을 우측 영상에서 찾는 과정이며, 이를 통해 스테레오 영상 내에서 거리 정보를 복원할 수 있다. 하지만 영상 내 특징이 불분명한 부분의 경우, 스테레오 매칭의 신뢰도가 낮기 때문에 Bad Pixel이 발생하게 된다. 제안하는 방법에서는 Bad Pixel을 보정하기 위해서 각 픽셀의 연관성을 고려하고자 한다. 일반적으로 동일한 물체는 비슷한 색상과 명도를 가진다. 따라서 컬러와 명도의 다중 임계값에 의해 각각 분할된 영역을 통해 영역 간의 연관성을 고려하여, 동일한 물체로 판단되는 부분을 재분할한다. 이후 분할된 픽셀들의 관계 정보에 따라 디스패리티 맵의 Bad Pixel을 보정하였다. 실험 결과, 제안하는 방법을 통해 기존 방법의 결과에서 Bad Pixel이 28% 감소함을 확인하였다.

This paper proposed the method to improve performance of a pixel, which has low accuracy, by applying image segmentation methods based on color and multi-threshold of brightness. Stereo matching is the process to find the corresponding point on the right image with the point on the left image. For this process, distance(depth) information in stereo images is calculated. However, in the case of a region which has textureless, stereo matching has low accuracy and bad pixels occur on the disparity map. In the proposed method, the relationship between adjacent pixels is considered for compensating bad pixels. Generally, the object has similar color and brightness. Therefore, by considering the relationship between regions based on segmented regions by means of color and multi-threshold of brightness respectively, the region which is considered as parts of same object is re-segmented. According to relationship information of segmented sets of pixels, bad pixels in the disparity map are compensated efficiently. By applying the proposed method, the results show a decrease of nearly 28% in the number of bad pixels of the image applied the method which is established.

본 논문에서는 칼라공간상의 거리와 이웃정보를 이용한 클러스터링을 통한 칼라영상 분할 방법을 제안한다. 영상의 픽셀들을 이웃관계를 유지하여 칼라공간으로 매핑한다. 칼라공간상에서 이웃하는 픽셀들을 클러스터링하여 영상의 세그먼트들을 찾는다. 클러스터링 방법으로서 인력을 모방하는 클러스터링(gravitational clustering)을 사용하였다. 이 방법으로 클러스터의 중심값과 클러스터 수를 미리 정해주지 않아도 자동적으로 결정할 수 있는 장점이 있다. gravitational 클러스터링에서 찾은 클러스터 수를 가지고 다른 클러스터링 방법에 입력으로 주어 결과를 비교해 본다. 본 논문에서는 이웃관계를 따라 클러스터링하는 것이 정확한 경계선을 찾는데 효과적임을 보여준다.