Earticle

영상 분할은 의료 영상 처리에서 중요한 역할을 수행하지만 영상 분할 알고리즘의 많은 연 산량은 실생활에서의 사용 을 제약하는 요인이 된다. 따라서 본 논문에서는 본 논문에서는 고해상도 의료 영상 분할 알 고리즘의 실시간 처리를 위해 GPGPU(general purpose graphics processing unit)상에서 영상 분할 알고리즘인 Fuzzy C-Means (FCM)의 병렬구현 방법을 제안한다. 본 논문에서는 GPU상에서 FCM 알고리즘의 고속화를 위해 대상 영상을 재 구성하고, 단일 쉐이더 코어에서 필요한 레지스터 개수를 바탕으로 FCM 알고리즘을 위한 최적의 쓰레드(thread) 와 블록(block)의 크기를 계산한다. 성능 평가를 위해 본 논문에서는 총 5가지 크기의 의료 영상(64×64, 128×128, 256×256, 512×512, 1,024×1,024)을 이용하여 단일 CPU대비 향상된 실행 시간을 측정하였고, 모 의실험 결과 64x64 영상을 제외한 나머지 영상에서 단일 CPU 보다 약 4배 높은 실행 시간 의 향상을 나타내었다.

Image segmentation plays a crucial role in numerous biomedical imaging applications. However, its high computational complexities require substantial amount of time and have limited its use in real applications. With this reason, this paper implements a fuzzy c-means (FCM) clustering algorithm on general purpose graphics processing unit (GPGPU) in order to run it in real time. This paper first reconfigures a target medical image for real-time processing of FCM, and calculates optimal sizes of threads and blocks for FCM based on the number of registers required by a shader core. To evaluate the proposed approach on GPGPU, this paper measures execution times to complete FCM on GPGPU by utilizing medical images at different resolutions (64×64, 128×128, 256×256, 512×512, 1,024×1,024). Experimental results indicate that execution times on GPGPU are approximately 4-fold faster than those on a single CPU for all of medical images except an 64×64 medical image.

This study proposed a novel framework for plant leaf disease identification. The proposed model consists of four steps including pre-processing, segmentation, feature extraction, and classification. At first, the unwanted noise and overfitting are removed, and also image contrast level is enhanced. Secondly, the Fuzzy C-Means (FCM) based Chameleon Swarm Algorithm (CSA) named as (FCM-CSA) is used for plant leaf diseased part segmentation. In the third stage, the feature extraction is performed using a fast GLCM feature extraction model. Finally, the Progressive Neural Architecture Search (PNAS) is used for plant leaf disease identification. The experimental investigations are carried out using MATLAB software with the Mendeley database. From this dataset, we have used Apple Cedar Apple Rust (ACAR), Cherry Powdery Mildew (CPM), Corn Common Rust (CCCR), Apple Healthy (AH), Grape Black Rot (GBR), Pepper Bell Bacterial Spot (PBBS), Potato Late Blight (PLB) and Tomato Leaf Mold (TLM) disease images. Different measures such as precision, recall, sensitivity, specificity, and accuracy results are used to validate the performance of the proposed model.

Two-dimensional polyacrylamide gel electrophoresis of proteins is a robust and reproducible technique. It is the most widely used separation tool in proteomics. Current efforts in the field are directed at development of tools for expanding the range of proteins accessible with two-dimensional gels. Proteomics was built around the two-dimensional gel. The idea that multiple proteins can be analyzed in parallel grew from two-dimensional gel maps. Proteomics researchers needed to identify interested protein spots by examining the gel. This is time consuming, labor extensive and error prone. It is desired that the computer can analyze the proteins automatically by first detecting then quantifying the protein spots in the 2-D gel images. In our previous work, we presented a new technique for segmentation of 2-D gel images using the fuzzy c-means algorithm using the notion of fuzzy relations. In this paper, we will describe the new relational fuzzy c-means algorithm (RFCM) and use it for automatic protein spots quantification. We will also use two methods to evaluate its performance: the unsupervised evaluation method and comparison with the expert spots quantification.

Aiming at partitioning an image into homogeneous and meaningful regions, automatic image segmentation is a fundamental but challenging problem in computer vision. It is well known that Fuzzy c-means (FCM) algorithm is one of the most popular methods for image segmentation. However, the FCM-based image segmentation algorithm must be manually estimated to determine cluster number by users. In this paper, we propose a novel cluster number adaptive fuzzy c-means image segmentation algorithm (CNAFCM) for automatically grouping the pixels of an image into different homogeneous regions when the cluster number is not known beforehand. We utilize the Grey Level Co-occurrence Matrix (GLCM) feature extracted at the image block level instead of at the pixel level to estimate the cluster number, which is used as initialization parameter of the following FCM clustering to endow the novel segmentation algorithm adaptively. We cluster image pixels according to their corresponding Gabor feature vectors to improve the compactness of the clusters and form final homogeneous regions. Experimental results show that proposed CNAFCM algorithm not only can spontaneously estimate the appropriate number of clusters but also can get better segmentation quality, in compare with those FCM-based segmentation methods recently proposed in the literature.

Aiming at the problem of traditional fuzzy C-means clustering algorithm that it is sensitive to the initial clustering centers and easy to fall into the local optimization, an improved algorithm that combines Improved Quantum Genetic Optimization with FCM algorithm is proposed. In this study, chromosomes are comprised of quantum bits encoded by real number. Chromosomes are renovated by quantum rotating gates and mutated by quantum hadamard gate. The gradients of object function are utilized in adjusting the value of rotating angle by a dynamic strategy. Each chain of genes represents a optimization result, Therefore, a double searching space is acquired for the same number of chromosomes. Experimental results show that the proposed method improves the stability and the accuracy of classification．

A design methodology of interval type-2 fuzzy c-means clustering algorithm-based fuzzy inference systems (IT2FCMFIS) is introduced in this paper. An interval type-2 fuzzy c-means (IT2FCM) clustering algorithm is developed to generate the fuzzy rules in the form of the scatter partition of input space. And the individual partitioned spaces describe the fuzzy rules equal to the number of clusters. The consequence part of the rule is represented by polynomial functions with interval set. To optimally construct of fuzzy model we exploit real-coded genetic algorithms with successive optimization. The proposed model is evaluated through the numeric experimentation.

An improved fuzzy c-means algorithm is put forward and applied to deal with meteorological data on top of the traditional fuzzy c-means algorithm. The proposed algorithm improves the classical fuzzy c-means algorithm (FCM) by adopting a novel strategy for selecting the initial cluster centers, to solve the problem that the traditional fuzzy c-means (FCM) clustering algorithm has difficulty in selecting the initial cluster centers. Furthermore, this paper introduces the features and the mining process of the open source data mining platform WEKA, while it doesn’t implement the FCM algorithm. Considering this shortcoming of WEKA, we successfully implement the FCM algorithm and the advanced FCM algorithm taking advantage of the basic classes in WEKA. Finally, the experimental clustering results of meteorological data are given, which can exactly prove that our proposed algorithm will generate better clustering results than those of the K-Means algorithm and the traditional FCM algorithm.

This paper presents a new clustering algorithm named improved type-2 possibilistic fuzzy c-means (IT2PFCM) for fuzzy segmentation of magnetic resonance imaging, which combines the advantages of type 2 fuzzy set, the fuzzy c-means (FCM) and Possibilistic fuzzy c-means clustering (PFCM). First of all, the type 2 fuzzy is used to fuse the membership function of the two segmentation algorithms (FCM and PCM), the membership function is an interval distribution, the determined fuzzy values which are the outputs of the FCM and PCM. Secondly, the initialization of cluster center and the process of type-reduction are optimized in this algorithm, which can greatly reduce the calculation of IT2PFCM and accelerate the convergence of the algorithm. Finally, experimental results are given to show the effectives of proposed method in contrast to conventional FCM, PFCM and type 2 fuzzy c-means.

With the fast development of image processing technique, segmentation related issues have gained special attention in the research community. In this paper, an improved FCM combining mean shift algorithm is proposed to improve the segmentation visual effects and efficiency of traditional FCM. Initially, image segmentation into many small homogeneous area using the mean shift algorithm is conducted, segmentation and uniform area, rather than pixels as the new node. Then, image local entropy is adopted to describe the new nodes spatial and gray feature. Finally, an exponential function which is able to well simulate human nonlinear visual reaction was used to measure the similarity between the new node and the cluster center node. The experimental result shows the effectiveness and robustness of our proposed FCM, further potential research is also discussed.

In this paper, a new median filter switcher is presented for suppression of impulsive noise in gray-level images. The proposed filter is Modified Adaptive Center Weighted Median (MACWM) filter with an adjustable central weight obtained by partitioning the observation vector space. Dominant points of the proposed approach are partitioning of observation vector space using fuzzy c-means clustering method, training procedure using LMS algorithm and then applying the freezing weights of each block to test image. The exprimental results show better performance in the impulse noise reduction over standard images relative the median (MED) filter, the switching scheme I (SWM-I) filter, the signal dependent rank order mean (SD-ROM) filter, the tristate median (TSM) filter, the fast peer group filter (FPGF), the fuzzy median (FM) filter, the PFM filter and the adaptive center weighted median (ACWM) filter.

A portable electronic tongue (E-Tongue) system using an array of ion-selective electrode (ISE) and personal digital assistants (PDA) for recognizing and analyzing food and drink have been designed. By the employment of PDA, the complex algorithm such as fuzzy c-means algorithm (FCMA) could be used in E-Tongue, PUMA could iteratively solve the cluster centers of pre-determined standard patterns. And the membership between the standard patterns and unknown pattern could be analyzed easily by the present E-Tongue combined with PDA.

FCM(fuzzy c-means)은 일반적으로 영상 분할에서 좋은 성능을 보인다. 하지만 공간 정보를 사용하지 않는 일반적인 FCM 알고리즘은 낮은 대비의 영상, 경계선이 뚜렷하지 않은 영상, 잡음이 포함된 영상의 분할에는 좋지 않은 성능을 보인다. 이와 같은 문제를 해결하기 위해 본 논문에서는 3x3 크기의 윈도우를 이용하여 윈도우 내의 중심 픽셀과 주변 픽셀간의 거리 정보를 소속 함수에 추가한 개선된 공간적 퍼지 클러스터링 알고리즘을 제안한다. 본 논문에서는 분할 계수, 분할 엔트로피, Xie-Bdni 함수와 같은 클러스터링 검증 함수를 이용하여 FCM 기반의 다양한 클러스터링 알고리즘과 제안한 알고리즘과의 성능을 비교하였다. 성능 평가 결과 제안한 알고리즘이 기존의 FCM기반의 클러스터링 알고리즘보다 클러스터링 검증 함수에서 성능이 우수함을 확인 할 수 있었다.

Conventional fuzzy c-means (FCM) algorithms have achieved a good clustering performance. However, they do not fully utilize the spatial information in the image and this results in lower clustering performance for images that have low contrast, vague boundaries, and noises. To overcome this issue, we propose an enhanced spatial fuzzy c-means (ESFCM) algorithm that takes into account the influence of neighboring pixels on the center pixel by assigning weights to the neighbors in a $3{\times}3$ square window. To evaluate between the proposed ESFCM and various FCM based segmentation algorithms, we utilized clustering validity functions such as partition coefficient ($V_{pc}$), partition entropy ($V_{pe}$), and Xie-Bdni function ($V_{xb}$). Experimental results show that the proposed ESFCM outperforms other FCM based algorithms in terms of clustering validity functions.

Building detection is a fundamental and challenging task in satellite image analysis and has significant importance in the wide range of applications such as creation and update of maps and GIS database, urban monitoring, and planning application. This paper presents an approach for building detection in high resolution remotely sensed images. The proposed approach firstly segmented the pan-sharpened image by Automatic Histogram based Fuzzy C-Means (AHFCM) algorithm. Secondly the vegetation and shadow regions are eliminated by using Normalized Difference Vegetation Index (NDVI) and ratio map in HIS model respectively. Finally the buildings and non building regions are classified by using Principal Component Analysis (PCA) and outcomes are further refined by morphological operations.

Gustafson and Kessel proposed a modified fuzzy c-Means algorithm based of the Mahalanobis distance. Though the algorithm appears more natural through the use of a fuzzy covariance matrix, it needs to calculate determinants and inverses of the c-fuzzy scatter matrices. This paper proposes a fuzzy clustering algorithm using pseudo mahalanobis distance, which is more easy to use and flexible than the Gustafson and Kessel's fuzzy c-Means.

The fuzzy c-means (FCM) is a frequently utilized algorithm at present. Yet, the clustering quality and convergence rate of FCM are determined by the initial cluster centers, and so an improved FCM algorithm based on canopy cluster concept to quickly analyze the dataset has been proposed. Taking advantage of the canopy algorithm for its rapid acquisition of cluster centers, this algorithm regards the cluster results of canopy as the input. In this way, the convergence rate of the FCM algorithm is accelerated. Meanwhile, the MapReduce scheme of the proposed FCM algorithm is designed in a cloud environment. Experimental results demonstrate the hybrid canopy-FCM clustering algorithm processed by MapReduce be endowed with better clustering quality and higher operation speed.

자동차 부품의 측정 시스템은 현재 고가의 장비가 대부분이다. 본 논문에서는 저가의 장비를 구현하려고 하였다. 자동차의 부품은 여러가지가 있으나, 이 중에서 현재 공장에서 측정에 어려움을 겪고 있는 에어콘 스윗치인 마그네트 코일 하우징을 대상으로 하였다 특히 측정 대상이 크고, 카메라의 화소수가 40만 이하일 경우, 측정의 중요한 포인트는 화소수이기 때문에 이를 정확히 알아 내는데, FCM(Fuzzy C-means) 알고리듬이 좋은 결과를 주지만 속성 공간에서 유사성만을 고려하고, 공간영역에서 유사성은 고려되지 않기 때문에 FCM은 "equal evidence"와 "ignorance"를 구분하지 못한다. 이를 개선하기 위해서 FCM를 수정하여 먼저 FCM로 처리하고 하고 이를 바탕으로 PCM(Possibilistic C-means)를 사용하였다. 길이를 측정하기 위해서는 표준이 되는 정확한 자가 필요하지만 실재로는 획득하기가 용이 하지 않기 때문에 이미 공장에서 수작업하여 얻은 합격 제품의 화소수들의 평균치를 표준값으로 하고 이를 표준 길이로 하였다. 결과를 모니터에 보여주고, RSC-232 포트를 통하여 신호를 마이크로프로세서에 전달하여 제품의 양호(good), 분량(bad)을 판별하는 신호를 발생하게 하였다.

The measuring systems of auto-parts are most of greater part very expensive. This paper tries to study to make a low-cost measuring equipment. There's several kinds of parts in automobile. In this study, we take aircon-switch called magnet coil-housing as the object of measurements. The measurements of this product is currently in difficult situations at factory. In the case of the mesuring objects being big sizes and camera sensor having under 410000 pixels, the key point is the number of pixels not to be changed whenever the same object is measured under the same position. We modified and used fuzzy c-means algorithm to get mostly without the change of the numbers of pixels exactly. Also, the standardized ruler is necessary to measure the length of the object but it is not easy to get the precised ruler. Therefore, the standard length has been taken as the mean value of the pixels in the previous passed objects manually obtained at factory. The results are displayed on monitor and transferred these signals to the microprocessor through RSC-232 port to determine a good or bad of products.

본 논문에서는 Fussy C-Means (FCM) 알고리즘에 의해 계산된 퍼지 클러스터들에 대한 평가 인덱스를 제안한다. 제안된 인덱스는 퍼지 클러스터들간의 인접성(inter-cluster proximity)을 이용한다. 클러스터 인접성을 도입함으로써 클러스터간의 중첩 정도를 계산할 수 있다. 따라서, 인접성 값이 낮을수록 클러스터들은 공간에 잘 분포하게 됨을 알 수 있다. 다양한 데이터 집합에 대한 실험을 통해서 제안된 인덱스의 효율성과 신뢰성을 검증하였다.

본 논문에서는 비선형 블라인드 채널등화기의 구현을 위하여 개선된 퍼지 클러스터(Modified Fuzzy C-Means: MFCM) 알고리즘을 제안한다. 제안된 MFCM은 기존의 유클리디언 거리 값 대신 Bayesian Likelihood 목적함수(fitness function)를 이용하여 비선형 채널의 출력으로 수신된 데이터들로부터 최적의 채널 출력 상태값(optimal channel output states)을 추정한다. 이렇게 추정된 채널 출력 상태 값들로 비선형 채널의 이상적인 채널 상태(desired channel states) 벡터들을 구성하고 이를 Radial Basis Function(RBF) 등화기의 중심(center)으로 활용함으로써 송신된 데이터 심볼을 찾아낸다. 실험에서는 무작위 이진 신호에 가우스 노이즈를 추가한 데이터를 사용하여 하이브리드 유전자 알고리즘 (GA merged with simulated annealing (SA): GASA)과 그 성능을 비교 하였으며, 제안된 MFCM을 이용한 등화기가 GASA를 활용한 것 보다 상대적으로 정확도와 속도 면에서 우수함을 보였다.

In this paper, a modified Fuzzy C-Means (MFCM) algorithm is presented for nonlinear blind channel equalization. The proposed MFCM searches the optimal channel output states of a nonlinear channel from the received symbols, based on the Bayesian likelihood fitness function instead of a conventional Euclidean distance measure. Next, the desired channel states of a nonlinear channel are constructed with the elements of estimated channel output states, and placed at the center of a Radial Basis Function (RBF) equalizer to reconstruct transmitted symbols. In the simulations, binary signals are generated at random with Gaussian noise. The performance of the proposed method is compared with that of a hybrid genetic algorithm (GA merged with simulated annealing (SA): GASA), and the relatively high accuracy and fast searching speed are achieved.

수천 개의 유전자 발현 정보를 가지고 있는 DNA 마이크로어레이 기술의 발달로 대량의 생물정보를 빠른 시간 내에 분석하는 것이 가능하게 되었다. 유전자를 분석하는 방법 중 하나인 클러스터링 방법은 비슷한 기능을 가진 유전자들을 집단화시켜서 집단내의 유전자들의 기능을 밝히거나, 미지의 유전자를 분석하는데 이용되고 있다. 본 논문에서는 유전자 데이터를 분석하기 위한 퍼지 클러스터링 방법과 이를 효과적으로 검증할 수 있는 베이지안 검증 방법을 제안한다. 퍼지 c-means 알고리즘을 사용하여 클러스터를 생성하고, 클러스터 결과를 기존의 퍼지 클러스터 검증 방법들과 본 논문에서 제안하는 베이지안 검증 방법을 사용하여 비교 평가한다. 베이지안 검증 방법은 각 유전자의 클러스터 멤버쉽을 확률로 이용하여 각 클러스터에 속할 확률을 계산하고, 이 값을 가장 크게 해주는 클러스터 집단을 선택한다. 이 방법은 기존의 퍼지 클러스터 검증 방법들과는 달리 클러스터 수에 무관한 평가가 가능한 장점을 가지고 있다. Serum과 Yeast 데이터에 대한 실험 결과, 베이지안 검증 방법의 유용성을 확인할 수 있었다.