外文翻譯---matlab 在圖像邊緣檢測(cè)中的應(yīng)用

1、<p><b>　　英文資料翻譯</b></p><p>　　MATLAB application in image edge detection</p><p>　　MATLAB of the 1984 countries MathWorks company to market since, after 10 years of

2、 development, has become internationally recognized the best technology application software. MATLAB is not only a kind of direct, efficient computer language, and at the same time, a scientific computing platform, it fo

3、r data analysis and data visualization, algorithm and application development to provide the most core of math and advanced graphics tools. According to provide it with the more than 500 math and eng</p><p>

4、　　MATLAB software has very strong openness and adapt to sex. Keep the kernel in under the condition of invariable, MATLAB is in view of the different application subject of launch corresponding Toolbox (Toolbox), has now

5、 launched image processing Toolbox, signal processing Toolbox, wavelet Toolbox, neural network Toolbox and communication tools box, etc multiple disciplines special kit, which would place of different subjects research w

6、ork.</p><p>　　MATLAB image processing kit is by a series of support image processing function from the composition, the support of the image processing operation: geometric operation area of operation and op

7、eration; Linear filter and filter design; Transform (DCT transform); Image analysis and strengthened; Binary image manipulation, etc. Image processing tool kit function, the function can be divided into the following cat

8、egories: image display; Image file input and output; Geometric operation; Pixels statis</p><p>　　Edge detection this</p><p>　　Use computer image processing has two purposes: produce more suitabl

9、e for human observation and identification of the images; Hope can by the automatic computer image recognition and understanding.</p><p>　　No matter what kind of purpose to, image processing the key step is

10、to contain a variety of scenery of decomposition of image information. Decomposition of the end result is that break down into some has some kind of characteristics of the smallest components, known as the image of the y

11、uan. Relative to the whole image of speaking, this the yuan more easily to be rapid processing.</p><p>　　Image characteristics is to point to the image can be used as the sign of the field properties, it can

12、 be divided into the statistical features of the image and image visual, two types of levy. The statistical features of the image is to point to some people the characteristics of definition, through the transform to get

13、, such as image histogram, moments, spectrum, etc.; Image visual characteristics is refers to person visual sense can be directly by the natural features, such as the brightness </p><p>　　The image is the ba

14、sic characteristics of edge, the edge is to show its pixel grayscale around a step change order or roof of the collection of those changes pixels. It exists in target and background, goals and objectives, regional and re

15、gion, the yuan and the yuan between, therefore, it is the image segmentation dependent on the most important characteristic that the texture characteristics of important information sources and shape characteristics of t

16、he foundation, and the image of the textu</p><p>　　The edge of the image is reflected by gray not continuity. Classic edge extraction method is investigation of each pixel image in an area of the gray change

17、, use edge first or second order nearby directional derivative change rule, with simple method of edge detection, this method called edge detection method of local operators.</p><p>　　The type of edge can be

18、 divided into two types: (1) step representation sexual edge, it on both sides of the pixel gray value varies significantly different; (2) the roof edges, it is located in gray value from the change of increased to reduc

19、e the turning point. For order jump sexual edge, second order directional derivative in edge is zero cross; For the roof edges, second order directional derivative in edge take extreme value.</p><p>　　If a p

20、ixel fell in the image a certain object boundary, then its field will become a gray level with the change. The most useful to change two features is the rate of change and the gray direction, they are in the range of the

21、 gradient vector and the direction to said. Edge detection operator check every pixel grayscale rate fields and evaluation, and also include to determine the directions of the most use based on directional derivative dec

22、onvolution method for masking.</p><p>　　Digital image processing technique has been widely applied to the biomedical field, the use of computer image processing and analysis, and complete detection and recog

23、nition of cancer cells can help doctors make a diagnosis of tumor cancers. Need to be made in the identification of cancer cells, the quantitative results, the human eye is difficult to accurately complete such work, and

24、 the use of computer image processing to complete the analysis and identification of the microscopic images have </p><p>　　Cell edge detection is the cell area of ??the number of roundness and color, shape a

25、nd chromaticity calculation and the basis of the analysis their test results directly affect the analysis and diagnosis of the disease. Classical edge detection operators such as Sobel operator, Laplacian operator, each

26、pixel neighborhood of the image gray scale changes to detect the edge. Although these operators is simple, fast, but there are sensitive to noise, get isolated or in short sections of a continuous</p><p>　　E

27、dge detection of MATLAB</p><p>　　MATLAB image processing toolkit defines the edge () function is used to test the edge of gray image.</p><p>　　(1) BW = edge (I, "method"), returns and

28、I size binary image BW, including elements of 1 said is on the edge of the point, 0 means the edge points. Method for the following a string of:</p><p>　　1) soble: the default value, with derivative Sobel ed

29、ge detection approximate measure, to return to a maximum gradient edge;</p><p>　　2) prewitt: with the derivative prewitt approximate edge detection, a maximum gradient to return to edge;</p><p>

30、　　3) Roberts: with the derivative Roberts approximate edge detection margins, return to a maximum gradient edge;</p><p>　　4) the log: use the Laplace operation gaussian filter to I carry filtering, through t

31、he looking for 0 intersecting detection of edge;</p><p>　　5) zerocross: use the filter to designated I filter, looking for 0 intersecting detection of edge.</p><p>　　(2) BW = edge (I, "meth

32、od", thresh) with thresh designated sensitivity threshold value, rather than the edge of all not thresh are ignored.</p><p>　　(3) BW = edge (I, "method" thresh, direction, for soble and prewit

33、t method specified direction, direction for string, including horizontal level said direction; Vertical said to hang straight party; Both said the two directions (the default).</p><p>　　(4) BW = edge (I, 

34、9;log', thresh, log sigma), with sigma specified standard deviation.</p><p>　　(5) [BW, thresh] = edge (...), the return value of a function in fact have multiple (" BW "and" thresh ")

35、, but because the brace up with u said as a matrix, and so can be thought a return only parameters, which also shows the introduction of the concept of matrix MATLAB unity and superiority.</p><p><b>　　

36、Last word</b></p><p>　　MATLAB has strong image processing function, provide a simple function calls to realize many classic image processing method. Not only is the image edge detection, in transform d

37、omain processing, image enhancement, mathematics morphological processing, and other aspects of the study, MATLAB can greatly improve the efficiency rapidly in the study of new ideas.</p><p>　　MATLAB 在 圖 像 邊

38、 緣 檢 測(cè) 中 的 應(yīng) 用</p><p>　　MATLAB自1984年由國MathWorks公司推向市場(chǎng)以來,歷經(jīng)十幾年的發(fā)展,現(xiàn)已成為國際公認(rèn)的最優(yōu)秀的科技應(yīng)用軟件。MATLAB既是一種直觀、高效的計(jì)算機(jī)語言,同時(shí)又是一個(gè)科學(xué)計(jì)算平臺(tái),它為數(shù)據(jù)分析和數(shù)據(jù)可視化、算法和應(yīng)用程序開發(fā)提供了最核心的數(shù)學(xué)和高級(jí)圖形工具。根據(jù)它提供的500多個(gè)數(shù)學(xué)和工程函數(shù), 工程技術(shù)人員和科學(xué)工作者可以在它的集成環(huán)境中交互或編程以

39、完成各自的計(jì)算。 </p><p>　　MATLAB軟件具有很強(qiáng)的開放性和適應(yīng) 性。在保持內(nèi)核不變的情況下,MATLAB 可以針對(duì)不同的應(yīng)用學(xué)科推出相應(yīng)的工具箱(Toolbox),目前已經(jīng)推出了圖像處理工具箱、信號(hào)處理工具箱、小波工具箱、神經(jīng)網(wǎng)絡(luò)工具箱以及通信工具 箱等多個(gè)學(xué)科的專用工具箱,極大地方便了不同學(xué)科的研究工作。</p><p>　　MATLAB的圖像處理工具包是由一系列

40、支持圖像處理操作的函數(shù)組成的,所支持的圖像處理操作有:幾何操作區(qū)域操作和塊操作;線性濾波和濾波器設(shè)計(jì);變換(DCT變換);圖像分析和增強(qiáng);二值圖像操作等。圖像處理工具包的函數(shù),按功能可以分為以下幾類:圖像顯示;圖像文件輸入與輸出;幾何操作;像素值統(tǒng)計(jì);圖像分析與增強(qiáng);圖像濾波; 性二維濾波器設(shè)計(jì);圖像變換;領(lǐng)域和塊操作;二值圖像操作;顏色映射和顏色空間轉(zhuǎn)換;圖像類型和類型轉(zhuǎn)換;工具包參數(shù)獲取和設(shè)置等。與其他工具包一樣,用戶還可以根據(jù)需要

41、書寫自己的 函數(shù),以滿足特定的需要,也可以將這個(gè)工具包和信號(hào)處理工具包或小波工具包等其他工具包聯(lián)合起來使用。</p><p><b>　　邊緣檢測(cè)概述</b></p><p>　　利用計(jì)算機(jī)進(jìn)行圖像處理有兩個(gè)目的:產(chǎn)生更適合人類觀察和識(shí)別的圖像;希望能由計(jì)算機(jī)自動(dòng)識(shí)別和理解圖像。</p><p>　　無論為了哪種目的,圖像處理中關(guān)鍵的一步就是對(duì)

42、包含有大量各式各樣景物信息的圖像進(jìn)行分解。分解的最終結(jié)果是被分解成一些具有某種特征的最小成分,稱為圖像的基元。相對(duì)于整幅圖像來說,這種基元更容易被快速處理。</p><p>　　圖像的特征是指圖像場(chǎng)中可用作標(biāo)志的屬性,它可以分為圖像的統(tǒng)計(jì)特征和圖像的視覺特 征兩類。圖像的統(tǒng)計(jì)特征是指一些人為定義的特征，通過變換才能得到,如圖像的直方圖、矩、頻譜等;圖像的視覺特征是指人的視覺可直接感 受到的自然特征,如區(qū)域的亮度、

43、紋理或輪廓等。利用這兩類特征把圖像分解成一系列有意義的目標(biāo)或區(qū)域的過程稱為圖像的分割。</p><p>　　圖像最基本的特征是邊緣,所謂邊緣是指其周圍像素灰度有階躍變化或屋頂變化的那些像素的集合。它存在于目標(biāo)與背景、目標(biāo)與目標(biāo)、區(qū)域與區(qū)域、基元與基元之間,因此,它是圖像分割所依賴的最重要的特征,也是紋理特征的重要信息源和形狀特征的基礎(chǔ),而圖像的紋理形狀特征的提取又常常要依賴于圖像分割。圖像的邊緣提取也是圖像匹配的

44、基礎(chǔ),因?yàn)樗俏恢玫臉?biāo)志,對(duì)灰度的變化不敏感,可作為匹配的特征點(diǎn)。</p><p>　　圖像的邊緣是由灰度不連續(xù)性所反映的。經(jīng)典的邊緣提取方法是考察圖像的每個(gè)像素在某個(gè)區(qū)域內(nèi)灰度的變化,利用邊緣鄰近一階或二階方向?qū)?shù)變化規(guī)律,用簡(jiǎn)單的方法檢測(cè)邊緣,這種方法稱為邊緣檢測(cè)局部算子法。</p><p>　　邊緣的種類可以分為兩種：①階躍性邊緣,它兩邊的像素的灰度值有顯著的不同；②屋頂狀邊緣,它位

45、于灰度值從增加到減少的變化轉(zhuǎn)折點(diǎn)。對(duì)于階躍性邊緣，二階方向?qū)?shù)在邊緣處呈零交叉；對(duì)于屋頂狀邊緣，二階方向?qū)?shù)在邊緣處取極值。</p><p>　　如果一個(gè)像素落在圖像中某一個(gè)物體的邊界上,那么它的領(lǐng)域?qū)⒊蔀橐粋€(gè)灰度級(jí)的變化帶。對(duì)這種變化最有用的兩個(gè)特征是灰度的變化率和方向,它們分別以梯度向量的幅度和方向來表示。邊緣檢測(cè)算子檢查每個(gè)像素的領(lǐng)域并對(duì)灰度變化率進(jìn)行量化,也包括方向的確定,大多數(shù)使用基于方向?qū)?shù)掩模求卷積

46、的方法。</p><p>　　數(shù)字圖像處理技術(shù)已被廣泛應(yīng)用到生物醫(yī)學(xué)領(lǐng)域，運(yùn)用計(jì)算機(jī)對(duì)圖像進(jìn)行處理和分析，并進(jìn)一步完成癌細(xì)胞的檢測(cè)與識(shí)別，能有效的協(xié)助醫(yī)生對(duì)腫瘤癌癥做出診斷。在識(shí)別癌細(xì)胞時(shí)，需要做出定量的結(jié)果，人眼很難準(zhǔn)確的完成這類工作，而利用計(jì)算機(jī)圖像處理完成顯微圖像的分析和識(shí)別已經(jīng)取得了很大的進(jìn)展。近年來國內(nèi)外醫(yī)學(xué)圖像研究者對(duì)癌細(xì)胞的檢測(cè)識(shí)別提出了很多理論和方法，對(duì)癌細(xì)胞的診斷具有十分重要的意義和實(shí)踐價(jià)值。&

47、lt;/p><p>　　細(xì)胞邊緣的檢測(cè)是進(jìn)行細(xì)胞面積圓度個(gè)數(shù)和顏色等形態(tài)及色度學(xué)的計(jì)算和分析的基礎(chǔ)，其檢測(cè)結(jié)果直接影響病情的分析和診斷結(jié)果。經(jīng)典的邊緣檢測(cè)算子如Sobel算子，Laplacian算子等利用圖像的每個(gè)像素鄰域內(nèi)灰度的變化來檢測(cè)邊緣。雖然這些算子計(jì)算簡(jiǎn)單、速度較快，但存在對(duì)噪聲干擾敏感，得到孤立或分小段連續(xù)邊緣像素，重疊相鄰細(xì)胞邊緣等缺陷，而利用最佳閥值分割和輪廓提取相結(jié)合的方法進(jìn)行邊緣檢測(cè)，通過迭代算法

48、求得圖像分割的最佳閾值，再利用輪廓提取算法，挖去細(xì)胞內(nèi)部像素點(diǎn)，最后剩余部分圖像就是細(xì)胞的邊緣，改變了傳統(tǒng)邊緣檢測(cè)算法的處理順序，通過MATLAB編程實(shí)現(xiàn)后，實(shí)驗(yàn)結(jié)果表明能有效抑制噪聲干擾影響，同時(shí)能夠客觀地、正確地選取邊緣檢測(cè)的門限值，從而進(jìn)行精確的細(xì)胞邊緣檢測(cè)。</p><p>　　邊緣檢測(cè)的MATLAB 實(shí)現(xiàn)</p><p>　　MATLAB圖像處理工具包定義了edge( )函數(shù)用于

49、檢測(cè)灰度圖像的邊緣。</p><p>　　(1) BW=edge(I,‘method’），返回與I大小一樣的二進(jìn)制圖像BW，其中元素1表示的是邊緣上的點(diǎn)，0表示非邊緣點(diǎn)。method為下列字符串之一:</p><p>　　1)soble：缺省值，用導(dǎo)數(shù)的Sobel近似值檢測(cè)邊緣，梯度最大點(diǎn)返回邊緣；</p><p>　　2)prewitt：用導(dǎo)數(shù)的Prewitt近似

50、值檢測(cè)邊緣，梯度最大點(diǎn)返回邊緣；</p><p>　　3)roberts：用導(dǎo)數(shù)的Roberts近似值檢測(cè)邊緣，梯度最大點(diǎn)返回邊緣；</p><p>　　4)loG：使用高斯濾波器的拉普拉斯運(yùn)算對(duì)I進(jìn)行濾波,通過尋找0相交檢測(cè)邊緣；</p><p>　　5)zerocross：使用指定的濾波器對(duì)I濾波后，尋找0相交檢測(cè)邊緣。</p><p>

51、　　(2)BW=edge(I,‘method’,thresh)中用thresh指定靈敏度閾值，所有不強(qiáng)于thresh的邊緣都被忽略。</p><p>　　(3)BW=edge(I,‘method’,thresh,direction),對(duì)于soble和prewitt 方法指定方向,direction為字符串,其中horizontal表示水平方向； vertical表示垂直方向；both表示兩個(gè)方向(缺省值)。<

52、;/p><p>　　(4)BW=edge(I,‘log’,thresh, sigma)，用sigma指定標(biāo)準(zhǔn)偏差。</p><p>　　(5)[BW,thresh]=edge( … )，函數(shù)的返回值實(shí)際上有多個(gè)(“BW”和“thresh”），但由于用中括號(hào)括起表示為一個(gè)矩陣，所以又可認(rèn)為只有一個(gè)返回參數(shù)，這也體現(xiàn)了MATLAB引入矩陣概念的統(tǒng)一性和優(yōu)越性。</p><p&g