機(jī)械外文翻譯文獻(xiàn)翻譯--原型基于顏色的圖像檢索與matlab

1、<p><b>　　英文原文</b></p><p>　　Prototyping Color-based Image Retrieval with MATLAB</p><p><b>　　Abstrac</b></p><p>　　Content-based retrieval of (image) datab

2、ases has become more popular than before. Algorithm develop-ment for this purpose requires testing/simulation tools,but there are no suitable commercial tools on the market.</p><p>　　A simulation environment

3、 for retrieving images from database according histogram similarities is presented in this paper. This environment allows the use of different color spaces and numbers of bins. The algorithms are implemented with MATLAB.

4、 Each color system has its own m-files.</p><p>　　The phases of the software building process are pre-sented from system design to graphical user interface (GUI). The functionality is described with snapshots

5、 of GUI.</p><p>　　1. Introduction</p><p>　　Nowadays there are thousands or hundreds of thousands of digital images in an image database. If the user wants to find a suitable image for his/her pu

6、rposes, he/she has to go through the database until the correct image has been found or use a reference book or some “intelligent” program. Video on demand (VoD) services also requires an intelligent search system for en

7、d-users. VoD systems’ search methods differ slightly from image database’s methods.</p><p>　　A reference book is a suitable option, if the images are arranged with a useful method, for example: 1)categories:

8、 animals, flags, etc, 2) names (requires a good naming technique) or 3) dates. An experienced user can use these systems as well as textual searches (keywords have to be inserted in a database) efficiently. There are sit

9、uations when a multi-language system has to be used. There a language independent search system’s best properties can be utilized. A tool which is based on the images</p><p>　　In the MuVi-project [1] this ki

10、nd of tool is under construction. It will cover the properties presented above.Research work on content-based image retrieval has been done in [2 – 6]. The system, which is presented in this paper, is a simulation enviro

11、nment, where MuVi’s color content based retrieval has been developed and tested.</p><p>　　2. System development</p><p>　　MATLAB is an efficient program for vector and matrix data processing. It

12、contains ready functions for matrix manipulations and image visualization and allows a program to have modular structure. Because of these facts MATLAB has been chosen as prototyping software.</p><p>　　2.1 S

13、ystem design</p><p>　　Before any m-files have been written, the system designhas been done. A system design for the HSV (hue, saturation and value) color system based retrieval process is presented in Figure

14、 1. Similar design has been done for all used color systems.</p><p>　　Figure 1: Function chart for HSV color space with 27 bins histogram.</p><p>　　Tesths27 is the main function for this color s

15、ystem and this number of bins. It calls other functions(hs27read, dif_hsv and image_pos) when needed. Each</p><p>　　color system has a main function of its own and variable number (2 – 3) of sub-functions. I

16、f there is no need for color space conversion there are 2 functions,otherwise 3 functions on the first branch of the function chart.</p><p>　　The function call of the main function is: matches=tesths27(imag

17、en,directory,num)</p><p>　　The variable imagen specifies the query image’s name and path. The directory is a path of the image database and num is a desired number of retrieved images.</p><p>　　

18、2.2 Functions</p><p>　　At this moment there are functions implemented for four color spaces: HSV, L*a*b*, RGB and XYZ [7]. Each color space has from 2 to 4 implementations for different numbers of bins. Ther

19、e are altogether 14 main functions.</p><p>　　For some color systems it is possible to make these functions dynamic, i.e. dynamic histogram calculation. Every color system / bin combination requires its own h

20、istograms and these can be made only with an exhaustive method (pixel by pixel). Histogram calculation takes ½ - 5 minutes per image, each</p><p>　　approximately 320×240 pixels, depending on the co

21、mplexity of the color space on 150 MHz Pentium. Thus it is not reasonable to let the user select a bin number freely, especially in the case of large databases.</p><p>　　The functions have been named so that

22、 the names contain information of the color space used, the purpose of the functions and the number of used bins. Some functions, for example image_pos, have been used by many or all main functions and these functions ha

23、ve not been named as described above.</p><p>　　The main function checks, if the function call is correct. If the query image’s name doesn’t contain a path, the function assumes that the image is situated in

24、the database directory. In addition to this, the main function checks, if the query image already has a histogram in the currently used database. If the required histogram is not there, the image read (for example hs27r

25、ead) function is called. This function also normalizes pixel values and arranges image matrix data to a vector format. A</p><p>　　The histogram will then be saved into the database directory. If the histogra

26、m already exists there, the three previous steps will not be executed. Now the query image has been analyzed. Then the main function will go through all images in the database directory with an almost similar algorithm

27、as in the case of the query image. The difference is that now there will be a</p><p>　　histogram difference calculation between the query image’s and current image’s histogram. Finally the image_pos function

28、 will be used to put a query image and the desired number of best match images on the display.</p><p>　　2.3 Linking</p><p>　　It is not possible to use a program before the main function and sub-

29、functions are connected to each other. The main function will be called from the command line or through the graphical user interface, which will be presented later in this paper. In both cases the function call will con

30、tain the same arguments. For multi-level search purposes separate main functions have been implemented, but it is possible to utilize “normal” functions and add one parameter, where the best matches array can be</p>

31、;<p>　　The main function calls an image read function with the image’s name. The histogram will be returned to the main function. If a color space conversion is needed, the conversion function will be called from

32、the read function with r, g and b –vectors. The histogram will be returned to the calling function. Finally the histogram build function will be called with converted color vectors. This function returns a quantized hist

33、ogram, which will go through all functions until it achieves the main func</p><p>　　The main function calls the histogram difference function with two histogram vectors and will get a difference value as a r

34、esponse. The difference function uses Euclidean-distance calculation, but it can be easily changed to</p><p>　　another algorithm due to the modularity of the program. If the difference is smaller than larges

35、t difference on a best match table, the current result will be written over the last result on the best match table. After that the table is arranged again in an ascending order of distance. When all the images have been

36、 analyzed, the sorted best match table, the number of desired output images, the query image’s name, the search image’s path and the database</p><p>　　path are transferred to the image_pos function. These va

37、lues can be transferred into larger components (vectors/containers). Now the program works faster with several input arguments, because there is no need for</p><p>　　picking up variables from a container.<

38、;/p><p>　　2.4 Graphical user interface</p><p>　　The graphical user interface (GUI) is an important part of software development. The designing of the GUI have to solve the following problems: learn

39、ing time, speed of performance, rate of errors by users, retention over time, and subjective satisfaction [9]. This software is, at the moment, intended to be used only for testing purposes. The most important property o

40、f this software is that the results of different test queries can be seen quickly and the</p><p>　　results can be saved safely on a disk. Thus the visual layout is not as important as in case of a commercial

41、 software product.</p><p>　　In Figure 2 the first screen on GUI is presented. The purposes of the buttons, menus and other components will be presented later. If this software is developed into a commercial

42、product, the menu bar will be disabled in the future and the exit and help buttons will be added on the canvas.</p><p>　　Figure 2: GUI before the search image selection.</p><p>　　In Figure 3 the

43、 search screen is presented just before starting a search. The user is shown a search image,and in this way he/she can be sure that the search will be made with the correct image.</p><p>　　Figure 3: GUI just

44、 before running a query.</p><p>　　The results of the query will be presented on the screen in the format which is presented in Figure 6.</p><p>　　3. Using the software</p><p>　　The

45、first screen has already been presented in Figure 2. The user can choose from pop-up menus (see Figure 4), if the search is made with one a color system or as a multi-level search. In a one-level search a roughly quant

46、ized or a more accurate histogram is used in one loop (one color system).</p><p>　　Figure 4: Color system selection from a popup menu.</p><p>　　The second menu is disabled because a one-level se

47、arch is selected.</p><p>　　In a multi-level search two different color systems /histograms are used. During the first loop the roughly quantized histograms are used and during the second loop. </p>&l

48、t;p>　　the more accurate histograms are utilized for the best matches from the first loop. The color system on the second loop can be either the same as on the first loop or a different one. For queries with one-level

49、search the selection of a second color system is disabled. The user can select the number of retrieved images at the final stage. The software can be linked to many image databases and the user can select a database wher

50、e the query will be directed.</p><p>　　The user can select a search image either from the same database where the query will be directed to (default) or from any directory in his/her PC. The selection will b

51、e made with the file – open dialog, which is presented in</p><p>　　Figure 5. The form can be cleared with the “Reset” button. A query is executed with the “Search” button. Finally the results of the search w

52、ill appear on the screen in a separate window, as presented in Figure 6. Earlier [8] the software</p><p>　　opened each image in a separate window and evaluating/saving the results is more difficult than afte

53、r the improvement. In the top left top corner is the original query image. Below that image the best matches are presented in a descending order of similarity from left to right and from top to bottom. The user can selec

54、t suitable images for further use with the “Copy selected” or the</p><p>　　“Print selected” buttons. The “New search” button closes this form and goes back to the original search form. The “Search similar” b

55、utton executes a new search where a query histogram is composed of histograms of the selected</p><p>　　images. If the user has selected a larger number than 21 as “Number of matching images”, the best matche

56、s will be shown on multiple screens. The user can browse these pages with the “Previous page” and “Following page”</p><p><b>　　buttons.</b></p><p>　　Figure 5: The query image selecti

57、on dialog. The language of the dialog depends on the language of the operating system used.</p><p>　　Figure 6: The results of a query will be presented graphically.</p><p>　　4. Summary</p>

58、<p>　　The color content-based retrieval requires algorithms, which give visually correct results. Correctly working algorithms can not be chosen before simulations. The software presented in this paper is intended

59、 to be used</p><p>　　for testing purposes. Some operations will be implemented, if the software is developed into a commercial product. Some modifications are under</p><p>　　construction.</p&

60、gt;<p>　　This software has been used as a testing platform for histogram quantization tests. The modularity of this program makes it possible to take new algorithms as a part of the software in a short time. MATLA

61、B makes</p><p>　　quick prototyping possible. A possibility to save figures (search results) directly on a disk is a fulfillment of the program’s requirements. After the results have been analyzed visually, t

62、he best algorithms will be taken as a part of the final software.</p><p>　　5. Acknowledgements</p><p>　　This work has been founded by the European Union– ERDF, the Technology Development Centre

63、Tekes, Alma Media, the Helsinki Telephone Company, Nokia Research Center, the Satakunta High Technology</p><p>　　Foundation and Ulla Tuominen’s Foundation</p><p><b>　　中文譯文</b></p&

64、gt;<p>　　原型基于顏色的圖像檢索與MATLAB</p><p>　　· 摘要</p><p>　　基于內(nèi)容的檢索數(shù)據(jù)庫(kù)(圖像)已經(jīng)變得越來越受歡迎。為了達(dá)到這一目的，需要發(fā)展算法檢測(cè)/模擬工具，但市場(chǎng)上沒有合適的商業(yè)工具。</p><p>　　本文介紹了一個(gè)模擬環(huán)境，能夠從數(shù)據(jù)庫(kù)

65、中檢索圖像直方圖的相似之處。該環(huán)境允許使用不同的色彩空間及柱，通過MATLAB實(shí)現(xiàn)算法。每一種顏色體系都有自己的m-files文件。</p><p>　　這個(gè)階段的軟件建設(shè)過程是從系統(tǒng)出發(fā)設(shè)計(jì)的pre-sented圖形用戶界面(GUI)，對(duì)GUI作了簡(jiǎn)短的功能描述。</p><p><b>　　1.介紹</b></p><p>　　現(xiàn)階段圖像數(shù)

66、據(jù)庫(kù)有成千上萬的數(shù)字圖像，如果用戶想要找一個(gè)適合他/她目的的圖象，他/她必須尋遍整個(gè)數(shù)據(jù)庫(kù)直到檢索出正確的圖片，或使用相關(guān)資料或者一些智能軟件。為了最終用戶，視頻點(diǎn)播服務(wù)也需要一個(gè)智能搜索系統(tǒng)。視頻點(diǎn)播系統(tǒng)的檢索方式與圖像數(shù)據(jù)庫(kù)檢索方式略有不同。</p><p>　　如果圖像采用一個(gè)有效的方法排列，一些相關(guān)資料是一個(gè)不錯(cuò)的選擇，例如：1）范疇：動(dòng)物、旗幟等等；2）名字（需要一個(gè)好的命名法）；3)日期：一位有經(jīng)驗(yàn)的

67、用戶可以如同文本檢索(關(guān)鍵詞必須插入一個(gè)數(shù)據(jù)庫(kù))一樣，有效地運(yùn)用這些系統(tǒng)。還有使用多語言系統(tǒng)的情況。一種基于圖象特性的工具可以是語言獨(dú)立的，一種獨(dú)立語言系統(tǒng)可以利用最佳屬性進(jìn)行搜索，這些屬性可以有顏色、形態(tài)、質(zhì)感、空間定位的形狀等。</p><p>　　在MuVi-project中，這種工具正在建設(shè)中，它涵蓋了以上各種性質(zhì)。對(duì)基于內(nèi)容的圖像檢索的研究工作,已經(jīng)在文獻(xiàn)[2 - 6]中完成了。該系統(tǒng)，文中已經(jīng)提到過，

68、是一種模擬環(huán)境，在它上面，基于顏色內(nèi)容檢索的MuVi已經(jīng)開發(fā)并測(cè)試了。</p><p><b>　　2. 系統(tǒng)開發(fā)</b></p><p>　　MATLAB是一種有效的矢量和矩陣的數(shù)據(jù)處理程序。它包含了完整的矩陣運(yùn)算函數(shù)和圖像可視化功能且允許程序有模塊化結(jié)構(gòu)。因?yàn)橐陨线@些因素，MATLAB被選為原型軟件。</p><p><b>　　

69、2.1系統(tǒng)設(shè)計(jì)</b></p><p>　　要在撰寫m-files之前,完成這個(gè)系統(tǒng)設(shè)計(jì)。HSV(色調(diào)、飽和度、亮度)系統(tǒng)設(shè)計(jì)，顏色系統(tǒng)檢索過程如圖1。類似的設(shè)計(jì)已為所有顏色系統(tǒng)使用。</p><p>　　圖1：HSV顏色空間及27個(gè)柱的流程圖。</p><p>　　Tesths27是這個(gè)顏色系統(tǒng)及這些柱的最主要功能。它可以根據(jù)需要調(diào)用其他功能(hs27

70、read,dif_hsv和image_pos)。每個(gè)顏色系統(tǒng)都有自己的主要功能和可變數(shù)目(2-3)的子功能。如果需要顏色空間轉(zhuǎn)換，在功能表的第一個(gè)分支上有兩個(gè)或者三個(gè)此功能。</p><p><b>　　主函數(shù)的調(diào)用功能：</b></p><p>　　matches=tesths27(imagen,directory,num)　　</p><p&g

71、t;　　imagen指定變量名稱及查詢圖像的路徑。</p><p>　　Directory是圖像數(shù)據(jù)庫(kù)的路徑,num是圖像檢索的預(yù)期數(shù)字。</p><p><b>　　2.2功能</b></p><p>　　到目前為止已實(shí)現(xiàn)了四個(gè)色彩空間:HSV、L*a*b*、RGB和 XYZ。每一種顏色空間已從2到4實(shí)現(xiàn)了不同數(shù)量的柱。共計(jì)14個(gè)主要功能。&

72、lt;/p><p>　　對(duì)于某些顏色系統(tǒng)使得這些動(dòng)態(tài)功能成為可能，即動(dòng)態(tài)直方圖計(jì)算。每個(gè)顏色系統(tǒng)/柱組需要依靠自身直方圖，這些可以僅用一個(gè)詳盡的方法(像素x像素)實(shí)現(xiàn)。直方圖計(jì)算每張圖片要花費(fèi)½- 5分鐘，每張大約320×240個(gè)像素，根據(jù)每150MHz 奔騰上，顏色空間的復(fù)雜性。因此讓每個(gè)用戶自由選擇每個(gè)柱的編號(hào)是不切實(shí)際的，特別是在大型數(shù)據(jù)庫(kù)。</p><p>　　這些

73、函數(shù)之所以如此命名，是因?yàn)檫@些名稱當(dāng)中包含了色彩空間的選擇，函數(shù)功能柱的數(shù)目。一些函數(shù),比如image_pos已經(jīng)應(yīng)用于許多或許所有的主函數(shù)當(dāng)中，并且這些函數(shù)并未按照上述方法命名。</p><p>　　主函數(shù)的功能測(cè)試，如果這個(gè)函數(shù)調(diào)用是正確的。如果這個(gè)查詢圖像的名字不包含路徑，該函數(shù)就會(huì)默認(rèn)圖像位于數(shù)據(jù)庫(kù)目錄下。除此之外，最主要的功能測(cè)試，查詢圖像的直方圖是否已經(jīng)位于當(dāng)前數(shù)據(jù)庫(kù)中。如果被查詢的直方圖不在數(shù)據(jù)庫(kù)中

74、，則調(diào)用圖像讀取(例如hs27read)函數(shù)。該函數(shù)還可以使像素值規(guī)格化，以及將圖像矩陣數(shù)據(jù)矢量化。該階段之后，將調(diào)用一個(gè)顏色空間轉(zhuǎn)換函數(shù)(如有需要)。最后調(diào)用一個(gè)量化函數(shù)根據(jù)正確數(shù)目的柱建立直方圖。</p><p>　　該直方圖將被保存到數(shù)據(jù)庫(kù)目錄下。如果直方圖已經(jīng)存在，之前的三步驟將不會(huì)被執(zhí)行?，F(xiàn)在對(duì)這個(gè)查詢圖像的已經(jīng)完成分析。然后主函數(shù)將依照查詢圖像，采用相似性算法，將圖像數(shù)據(jù)庫(kù)目錄下的所有圖像檢索一遍，不

75、同的是,現(xiàn)在將會(huì)在當(dāng)前的檢索圖像直方圖與查詢圖像直方圖之間有一個(gè)的差異計(jì)算。最后image_pos函數(shù)將查詢圖像以及檢索出來的要求數(shù)目的相似圖像顯示出來。</p><p><b>　　2.3鏈接</b></p><p>　　在主函數(shù)和子函數(shù)未鏈接之前要運(yùn)行一個(gè)程序是不可能的。主函數(shù)將通過命令行或通過圖形用戶界面調(diào)運(yùn),本文將后面陳述。這兩種函數(shù)調(diào)用的原理相同。對(duì)于多級(jí)搜

76、索功能分開的主要函數(shù)已經(jīng)實(shí)現(xiàn),但它可以利用普通程序并添加一個(gè)參數(shù),其中最匹配的數(shù)組可以轉(zhuǎn)為第二階段的比較函數(shù)。 </p><p>　　主要的函數(shù)通過圖像名稱調(diào)用圖像閱讀取函數(shù)，直方圖將返回到主函數(shù)。如果需要彩色空間轉(zhuǎn)換,轉(zhuǎn)換函數(shù)被讀取函數(shù)通過R、G、B向量調(diào)用。直方圖將返回到調(diào)用函數(shù)。最后,將通過轉(zhuǎn)換后的顏色向量，調(diào)用直方圖建立函數(shù)。該函數(shù)返回一個(gè)量化的直方圖，它將遍歷調(diào)用所有函數(shù)，直到實(shí)現(xiàn)主函數(shù)。</p&

77、gt;<p>　　主要函數(shù)根據(jù)兩個(gè)直方圖向量，調(diào)用直方圖差函數(shù)，并響應(yīng)一個(gè)得到的差值。差值函數(shù)使用歐氏距離計(jì)算,但它可以依據(jù)程序的模塊性很容易地更改為另一種算法。如果差值小于最佳匹配表中的最大差值,該結(jié)果將覆蓋最佳匹配表中的最后一條紀(jì)錄。最后，再將該表按照差值大小升序排列。當(dāng)所有的圖像分析完成后,最佳匹配的排序表,在所需的輸出圖像的數(shù)量,查詢圖像的名稱,檢索圖像的路徑和數(shù)據(jù)庫(kù)路徑均傳遞給image_pos函數(shù)。這些值可以傳

78、遞到更大的組件（向量/容器）?，F(xiàn)在，可以通過輸入幾個(gè)參數(shù)而快速實(shí)現(xiàn)，不需要從容器中查找變量。 </p><p><b>　　2.4圖形用戶界面</b></p><p>　　圖形用戶界面（GUI）是一個(gè)軟件開發(fā)的重要組成部分。該圖形用戶界面設(shè)計(jì)要解決以下問題：學(xué)習(xí)時(shí)間,速度的性能,用戶的錯(cuò)誤率,隨著時(shí)間的推移保留和主觀滿意度。該軟件目前僅預(yù)期用于測(cè)試目的，該軟件最重要的

79、性能是可以快速看到不同的測(cè)試查詢的結(jié)果,且結(jié)果可以安全地保存在磁盤上。因此,可視化設(shè)計(jì)倒不如商業(yè)軟件產(chǎn)品來得重要。</p><p>　　在圖2中呈現(xiàn)的屏幕是基于GUI的,按鈕、菜單和其他組件的功能將稍后介紹。如果這個(gè)軟件要開發(fā)成商業(yè)產(chǎn)品,菜單欄將被禁用,并將在畫布上添加退出和幫助按鈕。 </p><p>　　圖2:選擇查詢圖像之前的GUI。</p><p>　　圖3

80、查詢界面在開始查詢之前彈出。該用戶選擇一個(gè)查詢圖像,這樣他/她可以斷定,查詢到正確的圖像。</p><p>　　圖3:進(jìn)行查詢之前的GUI?！　?lt;/p><p>　　數(shù)據(jù)的結(jié)果將以圖像6中的界面顯示出來。</p><p><b>　　3. 軟件的使用</b></p><p>　　第一個(gè)界面已經(jīng)出現(xiàn)在圖2當(dāng)中。用戶可以從彈

81、出式菜單中選擇(見下圖),如果該系統(tǒng)是單色或多級(jí)檢索系統(tǒng)。在一個(gè)一級(jí)檢索系統(tǒng)中一個(gè)粗量化或更精確的直方圖從一次循環(huán)(一種顏色系統(tǒng))中得到。</p><p>　　圖4：系統(tǒng)顏色選擇從彈出式菜單當(dāng)中。 </p><p>　　菜單中的第二項(xiàng)被禁用,因?yàn)檫x擇了單級(jí)檢索。 </p><p>　　在一個(gè)多級(jí)檢索當(dāng)中，使用兩種不同顏色系統(tǒng)/顏色直方圖。在第一循環(huán)中，得到粗量化直方

82、圖，在第二重循環(huán)中，為了最佳匹配利用第一次循環(huán)從而得到更準(zhǔn)確直方圖。顏色系統(tǒng)的第二個(gè)循環(huán)可以與第一個(gè)循環(huán)相同或不同。對(duì)于一級(jí)檢索，二級(jí)顏色系統(tǒng)的選擇被禁用。用戶可以選擇圖像檢索在最后階段的數(shù)目。該軟件可以鏈接到許多圖像數(shù)據(jù)庫(kù),用戶可以選擇設(shè)定數(shù)據(jù)庫(kù)用來查詢。 用戶選擇一個(gè)查詢圖像,可以從查詢?cè)O(shè)定的同一（默認(rèn)）數(shù)據(jù)庫(kù)當(dāng)中或從他/她的PC下的任意目錄中，該選擇將以圖5中的文件-打開對(duì)話框出現(xiàn),該列表可以通過“重置”按鈕被清除，通過“檢索

83、”按鈕可以執(zhí)行檢索。最后,搜索結(jié)果將以一個(gè)獨(dú)立的窗口顯示在屏幕上的,如圖6。早些時(shí)候，該軟件將每一張圖都以一個(gè)獨(dú)立窗口打開，計(jì)算和保存結(jié)果都比改進(jìn)之后更困難。在頂部左上角是原來的查詢的圖像，在該圖像的下面是從左到右、從上到下，按照相似度降序排列的最佳匹配圖像。用戶可以進(jìn)一步使用“復(fù)制選擇”或 “打印選擇”按鈕選擇合適的圖像。 “新檢索”按鈕將關(guān)閉此列表,并又回到了原來的搜索列表。 “相似檢索”按鈕執(zhí)行一個(gè)新的檢索在查詢直方圖是由所有選定

84、圖像的直方圖組成時(shí)。如果用戶選擇了更大</p><p>　　圖5：選擇查詢圖片對(duì)話框。對(duì)話框的語言決于操作系統(tǒng)使用的語言。 </p><p>　　圖6：查詢結(jié)果以列表方式。</p><p><b>　　4. 摘要 </b></p><p>　　基于顏色內(nèi)容的檢索需要能提供直觀正確結(jié)果的算法，在模擬前不能選擇正確運(yùn)算的算法

85、。本文介紹的軟件以測(cè)試為目的的。一些功能還能夠擴(kuò)展，假如該軟件想要開發(fā)成商業(yè)產(chǎn)品，一些修改正在逐步實(shí)施過程。該軟件已被用來作為直方圖量化測(cè)試的測(cè)試平臺(tái)。該軟件的模塊化使得在新算法能夠在很短的時(shí)間內(nèi)成為軟件的一部分。MATLAB的使得快速原型的建造成為可能，使得直接在磁盤上快速保存數(shù)據(jù)（檢索結(jié)果），實(shí)現(xiàn)了軟件的要求。在結(jié)果可視化分析完成之后,最佳算法將成為這個(gè)最終軟件的一部分。 5. 鳴謝 </p><p>