哪位大神帮我用英语翻译一下这段话呗 急急急！！！！不要翻译器的 分可以多加的
图像分割是图像处理和计算机视觉的基本问题之一，是图像处理和图像分析的关键步骤。因此，对图像分割的研究一直是图像技术研究中的中心和热点。
本文对基于区域的图像分割的方法进行了综述，简单介绍了阈值法、区域生长和分裂合并法、基于统计学的算法等三种方法，并从中选取了最大熵算法，模糊C-均值聚类算法，分水岭算法，在matlab环境中对同一图像进行分割，进行重点研究。最大熵阈值法利用图像的灰度信息和梯度信息将图像的灰度直方图分成两个或多个独立的类，使得熵的总值最大，由此使得熵的最大灰度为最佳分割门限。fcm算法通过迭代来优化用于表示图像像素点与C类中心相似度的目标函数，以获取极大值来获取最优聚类。分水岭算法则假设二维图像是一个二维地形图，图像的灰度表示地形的高度，分水岭变换就是模拟该地形图侵入水的过程。本文通过不同方法的比较得到了最好的分割效果，仿真结果表明改进的算法大大提高了分割速度，并且保证了分割精度和区域完整性。
关键词：图像分割；matlab；分水岭变换；fcm算法；最大熵阈值
图像分割是图像处理和计算机视觉的基本问题之一，是图像处理和图像分析的关键步骤。因此，对图像分割的研究一直是图像技术研究中的中心和热点。
Image segmentation is one of the basic problems in image processing and computer vision, is a key step in image processing and analysis. Therefore, the study on the image segmentation has always been the focus and hot research in image processing.
本文对基于区域的图像分割的方法进行了综述，简单介绍了阈值法、区域生长和分裂合并法、基于统计学的算法等三种方法，并从中选取了最大熵算法，模糊C-均值聚类算法，分水岭算法，在matlab环境中对同一图像进行分割，进行重点研究。最大熵阈值法利用图像的灰度信息和梯度信息将图像的灰度直方图分成两个或多个独立的类，使得熵的总值最大，由此使得熵的最大灰度为最佳分割门限。fcm算法通过迭代来优化用于表示图像像素点与C类中心相似度的目标函数，以获取极大值来获取最优聚类。分水岭算法则假设二维图像是一个二维地形图，图像的灰度表示地形的高度，分水岭变换就是模拟该地形图侵入水的过程。本文通过不同方法的比较得到了最好的分割效果，仿真结果表明改进的算法大大提高了分割速度，并且保证了分割精度和区域完整性。
This paper reviewed the methods for region-based image segmentation, introduces the threshold method, region growing and splitting and merging, statistical algorithm based on three kinds of methods, and selected the maximum entropy algorithm from, fuzzy C- means clustering algorithm, watershed algorithm, in the MATLAB environment for the same image segmentation, focus on. The maximum entropy threshold method using image gray information and gradient information of the image histogram is divided into two or more independent classes, the maximum entropy value, which makes the maximum entropy is the optimum segmentation threshold. FCM algorithm to optimize the objective function for the representation of image pixels and the C class center similarity through iteration, in order to obtain maximum value to obtain the optimal clustering. Watershed algorithm assumes that two-dimensional image is a 2D topographic map, image gray represents the height of the terrain, the watershed transform is the simulation of the process of topographic map into the water. In this paper, through the comparison of different methods to get the best segmentation results, the simulation results show that the improved algorithm greatly improves the speed of segmentation, and ensure the accuracy of segmentation and regional integrity.
Image segmentation is one of the basic problems in image processing and computer vision, is a key step in image processing and analysis. Therefore, the study on the image segmentation has always been the focus and hot research in image processing.
本文对基于区域的图像分割的方法进行了综述，简单介绍了阈值法、区域生长和分裂合并法、基于统计学的算法等三种方法，并从中选取了最大熵算法，模糊C-均值聚类算法，分水岭算法，在matlab环境中对同一图像进行分割，进行重点研究。最大熵阈值法利用图像的灰度信息和梯度信息将图像的灰度直方图分成两个或多个独立的类，使得熵的总值最大，由此使得熵的最大灰度为最佳分割门限。fcm算法通过迭代来优化用于表示图像像素点与C类中心相似度的目标函数，以获取极大值来获取最优聚类。分水岭算法则假设二维图像是一个二维地形图，图像的灰度表示地形的高度，分水岭变换就是模拟该地形图侵入水的过程。本文通过不同方法的比较得到了最好的分割效果，仿真结果表明改进的算法大大提高了分割速度，并且保证了分割精度和区域完整性。
This paper reviewed the methods for region-based image segmentation, introduces the threshold method, region growing and splitting and merging, statistical algorithm based on three kinds of methods, and selected the maximum entropy algorithm from, fuzzy C- means clustering algorithm, watershed algorithm, in the MATLAB environment for the same image segmentation, focus on. The maximum entropy threshold method using image gray information and gradient information of the image histogram is divided into two or more independent classes, the maximum entropy value, which makes the maximum entropy is the optimum segmentation threshold. FCM algorithm to optimize the objective function for the representation of image pixels and the C class center similarity through iteration, in order to obtain maximum value to obtain the optimal clustering. Watershed algorithm assumes that two-dimensional image is a 2D topographic map, image gray represents the height of the terrain, the watershed transform is the simulation of the process of topographic map into the water. In this paper, through the comparison of different methods to get the best segmentation results, the simulation results show that the improved algorithm greatly improves the speed of segmentation, and ensure the accuracy of segmentation and regional integrity.
温馨提示：内容为网友见解，仅供参考
第1个回答 2013-05-27
Image segmentation is one of the basic problems in image processing and computer vision, is a key step in image processing and analysis. Therefore, the study on the image segmentation has always been the focus and hot research in image processing.
This paper reviewed the methods for region-based image segmentation, introduces the threshold method, region growing and splitting and merging, statistical algorithm based on three kinds of methods, and selected the maximum entropy algorithm from, fuzzy C- means clustering algorithm, watershed algorithm, in the MATLAB environment for the same image segmentation, focus on. The maximum entropy threshold method using image gray information and gradient information of the image histogram is divided into two or more independent classes, the maximum entropy value, which makes the maximum entropy is the optimum segmentation threshold. FCM algorithm to optimize the objective function for the representation of image pixels and the C class center similarity through iteration, in order to obtain maximum value to obtain the optimal clustering. Watershed algorithm assumes that two-dimensional image is a 2D topographic map, image gray represents the height of the terrain, the watershed transform is the simulation of the process of topographic map into the water. In this paper, through the comparison of different methods to get the best segmentation results, the simulation results show that the improved algorithm greatly improves the speed of segmentation, and ensure the accuracy of segmentation and regional integrity.
This paper reviewed the methods for region-based image segmentation, introduces the threshold method, region growing and splitting and merging, statistical algorithm based on three kinds of methods, and selected the maximum entropy algorithm from, fuzzy C- means clustering algorithm, watershed algorithm, in the MATLAB environment for the same image segmentation, focus on. The maximum entropy threshold method using image gray information and gradient information of the image histogram is divided into two or more independent classes, the maximum entropy value, which makes the maximum entropy is the optimum segmentation threshold. FCM algorithm to optimize the objective function for the representation of image pixels and the C class center similarity through iteration, in order to obtain maximum value to obtain the optimal clustering. Watershed algorithm assumes that two-dimensional image is a 2D topographic map, image gray represents the height of the terrain, the watershed transform is the simulation of the process of topographic map into the water. In this paper, through the comparison of different methods to get the best segmentation results, the simulation results show that the improved algorithm greatly improves the speed of segmentation, and ensure the accuracy of segmentation and regional integrity.
第2个回答 2013-05-27
China's journalism from the initial stage entered a stage of gradual improvement of the continuous deepening of the information era, the mainstream media more and more the tremendous impact of online media, "News Network" has a huge impact remainsthe leading position of the TV news media, "News Network" with some particularity. This article from the news content, the host to start a comprehensive overview of the main features of "Network News" to launch the three decades since.Which news content as the main target, the focus will be the choice of news content and content arrangement of these two aspects of analysis; host from the image of the temperament of the Host, clothing, broadcast features analysis. This study was to compare different periods of the broadcast of "Network News" summarized its characteristics, the effective experience advertised to help improve the comprehensive strength of China's news media.
第3个回答 2013-05-27
Segmentation is one of the basic problems of image processing and computer vision, are key steps in image processing and image analysis. Therefore, the research has been the image of image segmentation technology research centers and hot spots. This paper, the method of image segmentation based on region an overview, brief threshold method, the method of growth and Division merged three methods, based on statistical algorithms, and select the maximum entropy algorithm fuzzy c-means clustering algorithm, watershed, dividing the same image in the MATLAB environment, focused studies. Maximum entropy thresholding method uses image grayscale information and gradient information is a histogram of the image into two or more separate class, make the highest entropy value, thereby making maximum entropy grayscale is best split threshold. FCM algorithm by iterative optimization is used to represent the image pixels are similar to class c central objective function to get Maxima to obtain optimal cluster. Watershed algorithm for two-dimensional image is assumed to be a two-dimensional map, image grayscale representation of terrain height and watershed transformation is the process of simulation of the topographic map into the water. Through the comparison of the different methods in this article had the best split effects, simulation results indicate that improved split algorithm greatly improves speed and ensure segmentation accuracy and integrity.
相似回答
