CN102376088A - System for quantizing similarity between images on computer - Google Patents

System for quantizing similarity between images on computer Download PDF

Info

Publication number
CN102376088A
CN102376088A CN201010260502XA CN201010260502A CN102376088A CN 102376088 A CN102376088 A CN 102376088A CN 201010260502X A CN201010260502X A CN 201010260502XA CN 201010260502 A CN201010260502 A CN 201010260502A CN 102376088 A CN102376088 A CN 102376088A
Authority
CN
China
Prior art keywords
image
color
similarity
portion
analysis
Prior art date
Application number
CN201010260502XA
Other languages
Chinese (zh)
Inventor
张建铭
李禹亮
Original Assignee
康博公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 康博公司 filed Critical 康博公司
Priority to CN201010260502XA priority Critical patent/CN102376088A/en
Publication of CN102376088A publication Critical patent/CN102376088A/en

Links

Abstract

The invention relates to a system for quantizing the similarity between images on a computer, in particular to a method and a system for quantizing the similarity of images, which seem to be similar. The method comprises the following steps of: providing a group of images, selecting an image as a basal image, and comparing the other images with the basal image; selecting one or more parts of the basal image for comparison, and calculating color composition of the areas; checking the color composition to quantize the similarity or difference between the images, and appointing a score reflecting the quantized similarity or difference; and displaying a result. By the system and the method, a website owner is allowed to check whether a webpage crossing different browsers has a defect or not; and through analysis, obvious errors are marked, and even the offset of a single pixel can be marked.

Description

用于在计算上量化图像间相似性的系统 Used on a computing system quantization similarity between the images

技术领域 FIELD

[0001] 本发明总体上涉及图像处理的领域,并且更具体地,涉及在计算上量化图像之间的相似性和/或差异性。 [0001] The present invention relates generally to the field of image processing, and more particularly, to quantify similarities and / or differences between the images in the calculation.

背景技术 Background technique

[0002] 用于人工比较成对图像的方法是本领域已知的,诸如由人在图像之间进行视觉上的比较,以确定它们相似或者不同的方式。 [0002] A method for manually paired comparison image are known in the art, such as for visual comparison between the images by a human, to determine how they are similar or different. 然而,人为错误会导致这种评估的不准确性,并且对于较大的图像集合来说,由人驱动的比较可能是耗时或不可行的。 However, human error can lead to inaccuracies in this assessment, and for larger image collections, the man driven by a comparison can be time consuming or not feasible. 用于比较图像的其他已知方案可以标识一个图像与下一个不同,但是没有提供用于量化相似性或差异性程度的手段。 Other known solutions for comparison image may identify a different image to the next, but does not provide for quantifying the degree of similarity or difference means.

[0003] 应用图像分析的一个领域是在用户(诸如网站所有者)检查网页跨多个浏览器的一致性时。 One area [0003] image analysis is the user (such as a website owner) check consistency across multiple web browsers. 这些测试可以包括对数千图像的比较,对于这种比较来说,传统的分析可能太过费时。 These tests may include a comparison of thousands of images, for this comparison, the traditional analysis may be too time-consuming.

发明内容 SUMMARY

[0004] 在各种实施方式中,本发明提供用于对看似相似的图像之间的相似性进行量化的方法和系统。 [0004] In various embodiments, the present invention provides for the similarity between seemingly similar to the image quantization methods and systems. 给定一组图像,选择一个图像作为基图像,其余图像将与该基图像进行比较。 Given a set of images and select an image as a base image, the image will be compared to the rest of the base image. 选择基图像的一个或多个部分以便比较,并且计算这些区域的颜色构成。 To compare one or more portions selected image group, and calculates a color constituting these regions. 继而,检查颜色构成以确定图像之间的相似性或差异性的程度,为其指派反映定量差异的得分。 Then, check the color composition to determine the degree of similarity or difference between images, assigning a score to reflect its quantitative differences. 显示结果;可以使用阈值相似性来修改该显示,以筛选出结果。 Display results; a similarity threshold may be used to modify the display to screen results.

[0005] 在此描述的方法允许例如网站所有者检查网页跨不同浏览器是否存在缺陷;分析不仅标识明显的错误,甚至还标识单个像素的偏移。 [0005] The method described here allows for example to check web site owners across different browsers for defects; analysis not only identifies obvious errors, and even identify a single pixel offsets.

[0006] 本说明书中的描述并非是穷举性的,特别地,根据附图、说明书和权利要求,多个附加特征对于本领域普通技术人员而言将是易见的。 Description [0006] This specification is not exhaustive, in particular, the drawings, specification and claims, a number of additional features to those of ordinary skill in the art will be readily apparent. 而且,应当注意,本发明说明书中使用的语言在原则上是出于可读和指示目的而选择的,而不是为了限制发明主题。 Further, it should be noted that the language used in the specification of the present invention, in principle and are readable for the purpose of indicating selection, and not to limit the inventive subject matter.

附图说明 BRIEF DESCRIPTION

[0007] 图1示出了用于比较图像间相似性的系统的一个实施方式的计算环境; [0007] FIG. 1 shows one embodiment of a computing environment of a system for the similarity between the comparison image;

[0008] 图2A是示出与本发明结合使用的典型计算机的高级别框图; [0008] FIG 2A is a high level block diagram illustrating a typical computer for use with the present invention;

[0009] 图2B是图像分析提供者的一个实施方式的框图; [0009] FIG. 2B is a block diagram of one embodiment of an image analysis providers;

[0010] 图3是示出用于比较图像间相似性的过程的一个实施方式的流程图; [0010] FIG. 3 is a flowchart illustrating a process embodiment similarity between the comparison image;

[0011] 图4是描绘图3过程的一个实施方式的概念图; [0011] FIG. 4 is a conceptual diagram depicting one embodiment of the process of Figure 3;

[0012] 图5A描绘图像比较系统的用户界面的一个实施方式; [0012] FIG 5A depicts an image comparison system according to an embodiment of the user interface;

[0013] 图5B描绘分析已经开始时图5A的用户界面的一个实施方式; One embodiment [0013] Figure 5B depicts the analysis of the user interface has been started in FIG. 5A;

[0014] 图5C描绘分析已经完成时图5A的用户界面的一个实施方式; One embodiment [0014] Figure 5C depicts the analysis of the user interface of FIG. 5A have been completed;

[0015] 图5D描绘应用筛选时图5A的用户界面的一个实施方式; [0015] Figure 5D depicts an embodiment applied when filtering user interface of FIG. 5A;

6[0016] 图5E描绘显示多个相似性级别时图5A的用户界面的一个实施方式;以及 6 [0016] Figure 5E depicts an embodiment of a display when a plurality of similar level of the user interface of FIG. 5A; and

[0017] 图6描绘部分选择窗口的一个实施方式。 [0017] FIG 6 depicts an embodiment of a portion of the selected window.

[0018] 通过下文讨论,本领域的技术人员将会认识到,可以在不脱离本发明原理的情况下,使用在此示出的结构和方法的备选实施方式。 [0018] By following discussion, those skilled in the art will recognize that, without departing from the principles of the present invention, in an alternative embodiment of this embodiment of the structures and methods illustrated.

具体实施方式 Detailed ways

[0019] 图1示出了用于比较图像间相似性的系统130的一个实施方式的计算环境100。 [0019] FIG. 1 shows an embodiment of a computing environment for the exemplary system embodiment similarity between the comparison image 130 100. 计算环境100包括用户设备110、网站所有者120、图像分析提供者130以及网络140。 The computing environment 100 includes a user device 110, the site owner 120, 130, and image analysis provider network 140. 出于说明目的,图1所示的系统100的实施方式包括单个网站所有者120和单个用户设备110。 For illustrative purposes, the system shown in FIG. 1 embodiment 100 includes a single site owners 120 and a single user device 110. 然而,在其他实施方式中,系统100可以包括更多的用户设备110和/或更多的网站所有者120。 However, in other embodiments, the system 100 may include more user devices 110 and / or 120 more of the site owner.

[0020] 用户设备110包括可以从用户接收输入并且可以经由网络140来发射和接收数据的计算设备。 [0020] User device 110 may include receiving input from a user and may be transmitted via a network 140 and a computing device to receive data. 例如,用户设备110可以是桌面计算机、膝上型计算机、智能电话、个人数字助理(PDA)或者包括计算功能和数据通信能力的任何其他设备。 For example, user device 110 may be a desktop computer, a laptop computer, a smart phone, a personal digital assistant (PDA) or any other device including computing functionality and data communication capabilities. 用户设备110配置用于经由网络140与网站所有者120和图像分析提供者130通信。 The user equipment 110 is configured to provide for communication via the site owner 120 network 140 and image analysis 130.

[0021] 用户设备110包括浏览器应用112,其可以是任何可购得的web浏览器,例如Internet Explorer、Mozilla Firefox、Safari、Opera 或Google Chrome。 [0021] User device 110 includes a browser application 112, which may be any commercially available web browser such as Internet Explorer, Mozilla Firefox, Safari, Opera or Google Chrome. 在一个实施方式中,用户设备110通过使用浏览器应用112处理例如与网站所有者120提供的网页122相关联的标记语言文档来显示内容。 In one embodiment, the application 112 processes the user equipment 110 by using a browser with markup language document, for example, web page 122 associated with the owner of the website 120 provided by the display content. 通过执行标记语言文档中所包括的指令,浏览器应用112使用标记语言文档所描述的格式或呈现来显示网页122。 By instructions performed markup language document included in the browser application 112 format described in the markup language document or pages 122 to display the presentation.

[0022] 网站所有者120包括一个或多个web服务器,其包含一个或多个网页122,这些网页使用网络140向用户设备110传送。 [0022] website owners 120 comprises one or more web servers, comprising one or more web pages 122, 140 transmits these pages using the network 110 to the user equipment. 如上所述,由网站所有者120提供的网页122包括标记语言文档,其标识内容并且包括指定所标识内容的格式或呈现的指令。 As described above, provided by the web site owner 122 includes a markup language document 120, which identifies the content and includes an instruction to specify the identified content format or presented.

[0023] 图像分析提供者130包括促进在此描述方法的一个或多个计算设备。 [0023] Image analysis provider 130 comprises one or more promoting method described herein, a computing device. 图像分析提供者130对例如不同浏览器112所绘制的网页122的图像集与基图像进行比较,以快速确定图像间的相似性,并且向诸如网站所有者120的用户显示相似性信息。 Image analysis provider 130, for example, the base set of images of different image browser 112 drawn web 122 is compared to quickly determine the similarity between images, and display information to a user, such as the similarity of the site owner 120. 图像分析提供者130提供在此描述的核心功能,并且将结合图2A和图2B进一步详述。 Image analysis provider 130 provides the core functionality described herein, in conjunction with FIGS. 2A and 2B and described in further detail in FIG.

[0024] 网络140可以包括使用有线和无线通信系统二者的局域网和/或广域网、因特网或一个或多个内部网络的任意结合。 [0024] Network 140 may include the use of any of both wired and wireless local area network communication systems and / or wide area network, the Internet, or one or more internal network binding.

[0025] 图2A是示出用于与本发明结合使用例如作为图像分析提供者130、网站所有者120和/或用户设备110的典型计算机200的高级别框图。 [0025] FIG. 2A is a diagram illustrating a high-level block diagram of a typical computer using an image analysis provider 130, site owners 120 and / or 200 of user device 110, for example, in conjunction with the present invention. 示出了耦合至总线204的处理器202。 It illustrates a processor 204 coupled to bus 202. 与总线204耦合的还有存储器206、存储设备208、键盘210、图形适配器212、指点设备214以及网络适配器216。 Also coupled to bus 204 and memory 206, storage device 208, a keyboard 210, a graphics adapter 212, a pointing device 214 and a network adapter 216. 显示器218耦合至图形适配器212。 Display 218 is coupled to the graphics adapter 212.

[0026] 处理器202可以是任何通用处理器。 [0026] The processor 202 may be any general-purpose processor. 在一个实施方式中,存储设备208是硬盘,但是也可以是能够存储数据的任何其他设备,诸如可写压缩盘(CD)或DVD或者固态存储器器件。 In one embodiment, the storage device 208 is a hard disk, but may be any other device capable of storing data, such as a writeable compact disk (CD) or DVD, or a solid-state memory devices. 存储器206例如可以是固件、只读存储器(ROM)、非易失性随机访问存储器(NVRAM)和/或RAM,并且保存供处理器202使用的指令和数据。 The memory 206 may be, for example, firmware, read only memory (ROM), non-volatile random access memory (NVRAM), and / or RAM, and holds instructions and data for processor 202 use. 指点设备214可以是鼠标、轨迹球或者其他类型的指点设备,并且与键盘210结合使用以便向计算机200输入数据。 Pointing device 214 may be a mouse, trackball, or other type of pointing device, and in conjunction with the keyboard 210 to input data to the computer 200. 图形适配器212在显示器218上显示图像和其他信息。 Graphics adapter 212 displays images and other information on the display 218. 网络适配器216将计算机200耦合至网络114。 Network adapter 216 couples the computer 200 to a network 114. [0027] 图1的实体使用的计算机200的类型可以根据实施方式以及实体所使用的处理能力而变化。 Entity type computer 200 [0027] FIG. 1 is used may vary depending upon the embodiment and the processing power used by the entity. 例如,作为移动设备(诸如,PDA)的用户110通常具有有限的处理能力、小显示器218并且可能缺少指点设备214。 For example, as a mobile device (such as, a PDA) users 110 typically has limited processing power, a small display 218 and pointing device 214 may be missing. 反之,网站所有者120可以包括一起工作的多个刀片服务器。 On the other hand, site owners 120 may include multiple blade servers working together.

[0028] 如本领域已知的,计算机200适于执行如结合图2B描述的计算机程序模块。 [0028] As known in the art, such as the computer 200 is adapted to execute computer program modules described in conjunction with FIG. 2B. 模块存储在存储设备208中,被载入存储器206,并由处理器202来执行。 Stored in the storage module 208, it is loaded into memory 206 by the processor 202 for execution.

[0029] 图2B是图像分析提供者130的一个实施方式的框图。 [0029] FIG. 2B is a block diagram of an image analysis provider 130 to one embodiment. 图2B中所示的图像分析提供者130的实施方式是包括web服务器220、图像模块230、颜色信息模块M0、比较模块250、计分模块沈0、用户接口270以及图像分析存储观0的计算机系统。 Image analysis provided by the embodiment 130 shown in FIG. 2B is a web server 220, an image module 230, a color information module M0, the comparison module 250, scoring module 0 Shen, a user interface 270, and computer image analysis of the stored View 0 system.

[0030] web服务器220经由网络140将图像分析提供者130链接至最终用户设备110和网站所有者120,并且是用于此目的的一种手段。 [0030] web server 220 analyzes the image network 140 via links 130 to provide the end-user devices 110 and website owners 120, and a means for this purpose. web服务器220服务和捕获网页以及其他与web有关的内容,诸如Java、Flash、XML等。 web server 220 and capture web pages and other services and web-related content, such as Java, Flash, XML and so on.

[0031] 图像模块230接收图像并准备图像以供分析,并且是用于此目的的一种手段。 [0031] module 230 receives the image and the image prepared for image analysis, and a means for this purpose. 图像模块230例如经由web服务器220接收图像,并且提供绘制该图像以供显示所需的信息。 The image module 220 receives an image 230, for example, via a web server, and provides information to draw the desired image for display. 图像模块230接收对基图像及其部分的选择以供分析,并且指定相同比较中的其他图像作为分析图像。 The image receiving module 230 and the selected portion of the image group for analysis, and specifies other images of the same image analysis as comparison.

[0032] 按照在此描述的各实施方式,部分选择可以是整个图像、图像中的选定部分或者排除了某些区块的图像的单个部分。 [0032] In accordance with various embodiments described herein, may be selected section selects a single part or exclude certain portions of the image block of the entire image, image. 例如在网页捕获的情况下,对图像部分的选择是有用的,因为很多网页具有即时改变的动态内容。 For example, in the case of capture pages, select part of an image is useful because many pages with dynamic content instantly changed. 然而,在某些应用中,相对于选择独立部分进行比较而言,抓取整个页面或者从整体中排除特定区块对于分析而言更为有用。 However, in some applications, choose a separate part relative to the comparison, to crawl the entire page or a particular block more useful for analysis purposes excluded from the whole. 例如,由于对应于不同浏览器类型的标题的差异,将标题排除在选定用于分析的部分之外是有用的,以便防止对于已知跨不同浏览器而有差别的图像部分人为地使相似性得分降低,而不是作为任何错误的结果。 For example, since the different browser types corresponding to the title of the difference, the title selected for the excluded part analysis is useful in order to prevent the differential and the known image portion across different browsers similar artificially score lower, rather than as a result of any errors.

[0033] 在分析无需非常精确的某些实施方式中,可以使用页面部分的粗略JPEG格式。 [0033] In a very precise analysis without certain embodiments, the JPEG format may be a coarse portion of the page. 然而,在需要较精确细节的实施方式中,在抓取页面进行分析时将会把使用的任何压缩考虑在内,以防止错误地降低这些页面的相似性得分。 Consider any compression, however, require more precise details of the embodiments, crawling the pages will be analyzed, including the use of in order to prevent erroneous reduce the similarity score these pages.

[0034] 颜色信息模块240确定来自基图像的选定部分以及分析图像中对应部分的颜色信息,并且是用于此目的的一种手段。 [0034] Color information module 240 determines color information corresponding to the selected portion from the base portion, and analyzes the image of the image, and a means for this purpose. 颜色信息模块240例如通过确定图像中每种颜色的像素数目,来量化选定部分中的每种颜色。 Color information module 240, for example, by determining the number of pixels in the image for each color, each quantized color of the selected portion.

[0035] 比较模块250在基图像与分析图像之间比较颜色信息,并且是用于此目的的一种手段。 [0035] Comparative color information comparison module 250 between the base image and the image analysis, and a means for this purpose. 在一个示例中,将基图像的每个部分中的每种颜色的像素数目与每个分析图像的对应部分中的相同颜色的像素数目进行比较,并且记录图像间的相似性。 In one example, the number of pixels of the same color corresponding to number of pixels of each portion of the base portion of the image in each color and each of the analysis of the image is compared, and the similarity between the recorded images. 在不同示例中,测量相同数目的像素,但是根据其来记录图像间的差异。 In a different example, measuring the same number of pixels, but the differences between the recorded images in accordance with. 在另一示例中,首先使用散列函数来查看基图像与分析图像是否相同,并且仅在其不相同的情况下进行像素级别分析。 In another example, the first to use a hash function to see whether the analysis of images of the same image group, and only at the pixel level analysis which are not the same.

[0036] 计分模块260将图像间的颜色信息转换为得分,这是用于此目的的一种手段。 [0036] The scoring module 260 converts the color information of the score between images, which is a means for this purpose. 在一个示例中,针对部分中的每种颜色确定颜色得分,继而将各颜色得分相加在一起得到部分得分。 In one example, for a portion of each color determines the color of the score, the score of each color in turn are summed together to give the score portion. 继而将每个部分的部分得分相加在一起,得到结合部分得分。 The turn portion of each of the score portion summed together to give the score binding portion. 为了确定分析图像的相似性得分,用结合部分得分除以结合部分中的像素总数,并且将结果乘以100。 To determine the similarity score analysis of the image, with a binding part of the score portion is divided by total number of pixels in the binding, and multiplying the result by 100. 该相似性得分被指派给相关联的分析图像,并且范围从0(不相似)到100(等同)。 The similarity score is assigned to the associated image analysis, and ranges from 0 (no similarity) to 100 (equivalent). 如果使用基于散列的两步过程,则向被确定为等同的图像指派最高相似性得分(即,100)。 If a two-step process based on a hash, the image equivalent to assigning the highest similarity score is determined to be (i.e., 100). 如果只有一个部分(例如,排除区块的整体),则部分得分与结合部分得分相同,其除以部分中的像素数目,而后乘以100。 If only a portion (e.g., to exclude the entire block), then the binding portion of the score portion and the same score, divided by the number of pixels in the portion, and then multiplied by 100.

[0037] 用户接口模块270显示基图像和分析图像以及它们之间比较的结果,并且是用于此目的的一种手段。 [0037] User interface module 270 displays an image based image analysis and comparison between them and the result, and a means for this purpose. 用户接口模块270提供结合图5A-图6描述的用户界面。 Providing a user interface module 270 described in conjunction with FIG. 5A- FIG. 6 is a user interface. 用户接口模块270提供以下显示:多个图像、某图像作为基图像的指定、为进行比较对基图像部分的选择、分析进度信息、分析结果、用于调整所显示的分析图像的滑块以及用户界面中的各显示区域。 The user interface module 270 provides the following displays: a plurality of images, an image as a designated image group, compared to the base image selection portion, analyze the progress information, the analysis result, the analysis for the slider to adjust the displayed image and a user each display area interface.

[0038] 图像分析存储280可以是关系数据库或者存储在此描述的方法所使用的数据的任何其他类型数据库;其存储在图像比较过程期间产生的各种数据,并且是用于此目的的一种手段。 [0038] Image analysis memory 280 may be any other type of database data stored in a relational database, or the method described herein is used; various data generated during the image comparison process is stored, and for this purpose is a means. 例如,存储一个或多个表,其包括图像信息、部分信息、比较信息以及得分信息。 For example, store one or more tables, including image information, the partial information, comparing information and scoring information.

[0039] 模块220-270负责协调按照本发明的方法执行的过程,然而,模块无需是离散的模块;示出的配置仅仅意在作为示例,并且其他配置也在本发明的范围之内。 [0039] module 220-270 is responsible for coordinating the process according to the present invention performs the method, however, need not be discrete modules of the module; configuration shown only intended to be within the scope of example, and that other configurations are also the present invention.

[0040] 该系统可以使用单个计算机来实现,或者使用计算机的网络来实现,包括基于云的计算机实现。 [0040] The system may be implemented using a single computer, computer network or be implemented using, including cloud-based computer-implemented. 计算机优选地是服务器类的计算机,其包括一个或多个高性能CPU、主存储器以及计算机可读的持久存储,并且运行诸如LINUX或其变体的操作系统。 The computer is preferably a server class computer, which includes one or more high-performance CPU, a main memory and a persistent store computer-readable, and running an operating system such as LINUX, or a variant thereof. 所描述的系统130的操作可以通过硬件来控制,或者通过安装在计算机存储中并由此类服务器的处理器执行以实现在此描述功能的计算机程序来控制。 Operating system 130 described may be controlled by hardware, or by such a server processor to execute a computer program to implement the functions described herein are controlled by a computer installed in the store. 系统130包括在此描述的操作所需的其他硬件元件,包括网络接口和协议、安全性系统、用于数据录入的输入设备以及用于显示、打印或其他数据呈现的输出设备;没有示出这些和其他传统组件,以避免混淆相关细节。 The system 130 includes other hardware components required for the operation described herein, including network interfaces and protocols, security systems, data entry input device used for displaying, printing or other presentation data output device; not shown in these and other conventional components, in order to avoid confusing details.

[0041] 如上所述,系统130包括多个“模块”,其表示用于提供指定功能的计算逻辑。 [0041] As described above, system 130 includes a plurality of "modules", which indicates the computation logic for providing the specified functions. 引擎可以通过硬件、固件和/或软件来实现。 Engine can be implemented in hardware, firmware and / or software. 模块有时等效地称为“引擎”或“服务器”。 Modules are sometimes equivalently referred to as "engine" or "server." 将会理解,命名的组件代表本发明的一个实施方式,并且其他实施方式可以包括其他组件。 It will be understood, named components representative of one embodiment of the invention, and other embodiments may include other components. 而且,其他实施方式可以缺少在此描述的组件和/或以不同的方式在组件之间分布描述的功能。 Further, other embodiments may lack components and / or functionality in varying ways described distributed between the components described herein. 此外,分配给不止一个组件的功能可以合并到单个组件中。 Further, the function assigned to more than one component may be combined into a single component. 在将这里描述的引擎实现为软件的情况下,引擎可以实现为单机程序,但是也可以通过其他手段来实现,例如,实现为较大程序的部分、多个独立的程序或者一个或多个静态或动态链接库。 In the engine described herein will be implemented as in the case of software, the engine may be implemented as a standalone program, but can also be achieved by other means, e.g., implemented as part of a larger program, or a plurality of separate programs or more static or dynamic link library. 在任何这些软件实现中,引擎存储在系统130的计算机可读持久存储设备上,被载入存储器,并由系统的计算机的一个或多个处理器来执行。 In any of these software implementations, the engine system 130 stored in a computer-readable on persistent storage device, is loaded into a memory of a computer system by one or more processors to perform.

[0042] 图3是示出用于比较图像间相似性的过程的一个实施方式的流程图。 [0042] FIG. 3 is a flowchart illustrating a process embodiment similarity between the comparison image. 在此描述的方法允许例如网站所有者检查网页跨不同的浏览器是否存在缺陷;分析不仅标识明显的错误,甚至还标识单个像素的偏移。 The method described here allows site owners such as checking pages across different browser for defects; analysis not only identifies obvious errors, and even identify a single pixel offsets.

[0043] 过程开始于系统接收305对基图像以及基图像的至少一个部分的选择。 Selecting at least a portion of the [0043] process begins group received 305 pairs of images and image groups in the system. 基图像是将被用作与一组分析图像的图像比较的基础,并且所述部分是将要比较的图像区域。 Group image is an image to be used as the basis for a set of comparative analysis of the image, and the portion is a region of the image to be compared. 在一个实施方式中,直接通过在图像上描画矩形来选择部分。 In one embodiment, the rectangular drawing directly on a selected portion of the image. 在另一实施方式中,部分是将图像的某些区块(例如,网页上的动态内容)排除在外的整个图像(例如,网页);在此示例中,通过在要从部分中排除的区块上描画矩形来间接地选择部分。 In another embodiment, the part of the certain block (e.g., dynamic content on a webpage) exclude images of the entire image (e.g., page); In this example, by excluding from the portion of the region drawing the rectangular block selection portion indirectly. 按照其他实施方式,部分具有其他形状和大小;在这里以及附图中使用矩形是为了描述的简便。 In other embodiments, portions having other shapes and sizes; used herein and in the drawings for simplicity rectangular description. 例如,按照一个实施方式,通过提供对应于选择区域(要包括,或者要排除)边界的坐标来选择部分。 For example, according to one embodiment, by providing the region corresponding to the selection (to include, or to exclude) the selected portion of the boundary coordinates. 按照另一实施方式,基图像是网页,并且部分是整个网页。 According to another embodiment, the base image is a page, the entire page and is part. 该列举意味着示例性而非排他性。 This list is meant as illustrative and not exclusive. 针对部分选择的图像区块可以基于部分内容的重要性来确定。 Part of importance can be determined based on the image for a block section selection. 例如,公司标志可能是网页提供者想要确保始终无误显示的区块。 For example, the company logo is a web provider may want to make sure the block is always correctly displayed. 类似地,排除的区块通常基于那些区块的内容,例如,具有动态内容的区块可能被排除。 Similarly, the blocks are commonly excluded based on the contents of those blocks, e.g., blocks with dynamic content may be excluded.

[0044] 接下来,系统确定310来自基图像的部分的颜色信息,这例如是通过量化基图像的选定部分中的每种颜色。 [0044] Next, the system 310 determines the color information from the base portion of the image, in this example by a group selected portion of the quantization of the image of each color. 确定310包括在每一次迭代中针对每个选定部分确定每种颜色的像素数目。 Determining 310 comprises determining the number of pixels of each color for each selected portion of each iteration.

[0045] 简要参考图4,其示出了图3的过程的简化示例的概念图。 [0045] Referring briefly to FIG. 4, which illustrates a simplified conceptual diagram of an example process of FIG. 3. 在基图像410中,选定了三个部分425、430、435。 In the base image 410, the selected three parts 425,430,435. 在此示例中,示出了三种颜色,分别由黑色方块(颜色A)、斜线方块(颜色B)和白色方块(颜色C)表示。 In this example, it shows three colors, respectively, shaded boxes (color B), and white squares (color C) represented by the black squares (color A). 如引出的框中所示,部分1425是8像素部分,并且包括2个颜色A的像素、3个颜色B的像素和3个颜色C的像素。 As shown in box drawn, part 1425 is a pixel portion 8, and includes a pixel two pixels of color A, color B, the pixel 3 and the three color C. 类似地,部分M30是6像素部分,包括4个颜色A的像素和2个颜色C的像素(没有颜色B的像素);部分3435是4像素部分,包括1个颜色A的像素、2个颜色B的像素和1个颜色C的像素。 Similarly, part M30 is a 6 pixel portion, comprising four pixels of color A and two color C of pixel (no color B pixels); part 3435 is 4 pixel portion includes one pixel of color A, two colors pixels B and C of a color.

[0046] 继续参考图3,系统继而接收315至少一个分析图像以便与基图像比较。 [0046] With continued reference to FIG. 3, the system 315 then receives at least one image analysis for comparison with the base image. 在一个示例中,基图像和分析图像对应于不同设备和/或不同浏览器所绘制的网页。 In one example, the base image analysis, and the page image corresponding to different devices and / or drawn by different browsers. 通常,接收多个图像以便与基图像比较。 Typically, receiving a plurality of images for comparison with the base image. 从分析图像中标识与基图像的选定部分对应的部分。 Analyzing the image portions selected from the portions of the base corresponding to the identification image.

[0047] 系统继而例如使用上文描述的过程来确定320来自分析图像的颜色信息。 [0047] The system can then, for example, using the procedures described above to determine the color information 320 from the analysis of the image. 再一次,在一次迭代中量化基图像的选定部分中的每种颜色,这例如是通过针对每个选定部分来确定每种颜色的像素数目。 Again, the quantization of the selected portion of the image group for each color in the first iteration, for example by determining the number of pixels of each color for each selected portion. 再次参考图4,分析图像420具有三个对应部分440、445、450。 Referring again to Figure 4, analyzing the image 420 with corresponding portions of three 440,445, 450. 部分1' 440是对应于部分1425的8像素部分,并且包括2个颜色A的像素、4个颜色B的像素以及2个颜色C的像素。 Part 1 '440 corresponding to the 8-pixel portion 1425 includes a pixel and two pixels of color A, color B, four and two pixels of color C. 类似地,部分2' 445是对应于部分M30的6像素部分,包括4个颜色A的像素和2个颜色C的像素(没有颜色B的像素)。 Similarly, part 2 '6 445 corresponding to the pixel portions of M30, four pixels including a pixel of color A and color C 2 (no B color pixels). 部分3' 450是对应于部分3435的4像素部分,包括1个颜色A的像素、1个颜色B的像素和2个颜色C的像素。 Part 3 '450 corresponding to the portion 3435 of four pixels, one pixel includes a pixel of color A, color B, a pixel C, and the two colors.

[0048] 上述步骤305-320不限于上文描述的顺序。 [0048] The sequence of steps 305-320 is not limited to the above description. 在备选实施方式中,接收305,315基图像和分析图像二者,从基图像中选择部分,继而确定310,320每个的颜色信息,作为比较步骤325的基础。 In an alternative embodiment, the group receiving both the images and the analyzed images 305, 315, from the base image selection portion, 310, 320 then determines for each color information, as a basis for the comparing step 325.

[0049] 再次返回图3,系统接下来在基图像与分析图像之间比较325颜色信息。 [0049] Returning to Figure 3, the system then compares again between the base 325 and the image analysis of the image color information. 在一个示例中,将基图像的每个部分中的每种颜色的量化颜色(例如,像素数目)与每个分析图像的对应部分中的量化颜色进行比较,并且标注颜色之间的相似性以及例如在基图像与分析图像之间共同的给定颜色的像素数目。 In one example, each quantized color of the base portion of each color image (e.g., number of pixels) is compared with each corresponding part of the analysis of the image color quantization, and the similarity between the label and the color for example, between the substrate and the image analysis of the image pixel colors common to a given number. 在不同的示例中,首先对来自基图像和分析图像的颜色量化信息使用散列函数,以查看其是否匹配,并且仅在其不匹配的情况下,进行更详细的(例如,像素级)相似性确定。 In various examples, the first quantization information used for the color analysis of the image from the base image and the hash function to see if it matches, and only in the case where it does not match, in more detail (e.g., pixel level) similar determination.

[0050] 尽管图3的过程是关于基图像与分析图像之间颜色信息的相似性来描述的,但是备选地,可以使用图像之间颜色信息的差异性来进行比较。 [0050] Although the processes of FIG. 3 is described with respect to the similarity of color information between the base image and the image analysis, but alternatively, may be used a color difference between the image information for comparison. 类似地,按照另一实施方式,基图像部分可以备选地包括排除选定区块的整个图像。 Similarly, according to another embodiment, the base image portions may alternatively comprise a selected exclude the entire image block.

[0051] 参考图4,将基图像410中部分1410的每个颜色的颜色信息与分析图像中的部分1' 440进行比较。 [0051] Referring to FIG 4, color information of the image 410 analyzes the image group of each color section 1410 in Part 1 of the '440 is compared. 在此示例中,比较信息包括有多少像素是相似的,如下表所示。 In this example, the information comprises comparing how many pixels are similar, as shown in the following table.

[0052] [0052]

Figure CN102376088AD00111

[0053] 表1 [0053] TABLE 1

[0054] 例如,对于基图像410的部分1425与分析图像420的部分1,440,颜色A具有2个像素的相似性(每个具有2个颜色A的像素),颜色B具有2个像素的相似性(部分1425具有3个像素,但是部分1,440只有2个;其间共同的颜色B的像素数目为幻,并且颜色C具有3个像素的相似性(部分1425具有3个像素,部分1'具有4个像素;其间共同的颜色C的像素数目为3)。针对部分2和部分3示出了类似信息。 [0054] For example, the base portion 1425 of image 410 and image analysis section 420 1,440, color A similarity two pixels (each pixel having two color A), the color of two pixels having B similarity (having three pixel portion 1425, but only two portions of 1,440; the number of pixels therebetween common color B is magic, and a color similarity C 3 pixels (a pixel portion 1425 having 3, part 1 'having four pixels; the number of pixels therebetween are common color C 3) for the portion 2 and portion 3 illustrate similar information.

[0055] 再次参考图3,系统接下来基于比较325来确定330图像间的相似性。 [0055] Referring again to FIG. 3, the system 325 then compares the determined based on the similarity between the images 330. 该确定330的一个示例使用如关于图4所述在步骤325中确定的相似性值。 An example of the determination 330 is used as the similarity value with respect to FIG. 4 is determined in step 325. 继续上文示例,对于每个分析图像,确定每种颜色的相似性(颜色得分),继而将其相加在一起得到部分得分(参见表1中的“每部分总计”列)。 Continuing the example above, for each image is analyzed to determine the similarity of each color (the color rating), which in turn are summed together to give the score portion (see Table 1, "each section Total" column). 继而将每个部分的部分得分相加,以得到结合部分得分(表1中“每部分总计”列的最后一行)。 The portion of each portion then scores are summed to obtain ( "Every part Total" column of the last row in Table 1) of binding score portion. 为了确定分析图像的相似性得分,将结合部分得分除以部分中的像素总数(表1中的“图像总计”行,“每部分总单元”列),并且将结果乘以100。 To determine the similarity score of the image analysis, the total number of pixels binding portion divided by the score section (in Table 1, "total image," OK "means every part of the total" column), and multiplying the result by 100. 例如,(16/18)*100 = 89。 For example, (16/18) * 100 = 89. 将该得分指派给相关联的分析图像。 The score assigned to the associated image analysis. 使用此计算,相似性得分的变化范围从0 (不相似)到100 (等同)。 Using this calculation, the similarity score ranges from 0 (no similarity) to 100 (equivalent).

[0056] 备选地,使用两步过程,其中首先如上所述针对基图像和分析图像的颜色量化信息进行散列,以确定图像是否等同。 [0056] Alternatively, a two-step process, wherein the first quantization information as described above for the color image and the analysis of the image group hashed to determine whether the image equivalent. 如果它们是等同的,则指派最高得分(例如,100)。 If they are identical, the highest score is assigned (e.g., 100). 如果它们不等同,则可以例如按照上文描述的过程来确定相似性得分。 If they are not identical, for example, it may be in accordance with the procedure described above to determine the similarity score.

[0057] 在另一备选中,测量图像之间差异的量,使得确定图像间的相似性包括:将具有最低差异得分的图像表征为最相似的。 [0057] In another alternative, the amount of the difference between the measurement image, so that the similarity between images is determined comprising: characterizing the image having the lowest score difference was most similar.

[0058] 上述过程对应于对单个分析图像的分析。 Analysis of a single analysis of the image [0058] corresponding to the above-described process. 然而,如在此讨论的,在多数情况下分析将应用于多个分析图像。 However, as discussed herein, the plurality of analysis of the image analysis is applied in most cases. 由此,过程中的另一步骤包括确定340附加的图像是否需要比较,如果需要,则针对附加的图像重复比较。 Thus, the process further comprises the step of determining whether additional images need to be compare 340, if desired, the comparison is repeated for the additional image. 上述过程可以促进对多个分析图像的快速比较。 The above process may facilitate a quick comparison of multiple image analysis. 基于散列的备选方案将允许在短时间内扩展至大量分析图像;只有那些被确定为不等同的图像才会进行完整的分析过程。 Hash-based alternatives will allow the extension of time to analyze a large number of images; only those who are determined to be not the same as the image analysis process will be complete. 如果有很多分析需要系统处理,则可以将提交分析的新图像置于队列中,例如以先进先出为基础操作的队列。 If there are many systems require analysis processing, you can submit a new image analysis in a queue, for example, FIFO queue based operation.

[0059] 上述过程期间产生的各种数据被存储。 Various data generated during [0059] The above-described process is stored. 例如,一个或多个数据库存储一个或多个表,其包括关于每个图像(基图像,分析图像)的信息、每个部分大小/形状/位置、每个个体比较、每个得分(中间得分或相似性得分)以及类似信息。 For example, one or more databases that store one or more tables that include information about each image (image group, the image analysis), each portion size / shape / position of each individual comparison, each of the score (score intermediate or similarity score), and the like.

[0060] 系统还例如在用户界面中显示350相似性信息,将结合图5A-图6详述。 [0060] The system 350 also displays similar information, for example, in the user interface, will be described in detail in conjunction with FIG. 5A- FIG. 在一个示例中,显示包括:要比较的图像、用于选择基图像部分的功能、分析过程的进度指示、每个分析图像的得分以及用于通过设置阈值相似性得分来限制所显示分析图像的功能。 In one example, the display comprising: an image to be compared, a functional group selected image portion, the analysis indicates the progress of the process, the analysis score for each image, and means for setting a threshold similarity score to restrict the displayed image analysis Features.

[0061] 图5A描绘了图像比较系统的用户界面500的一个实施方式。 [0061] FIG 5A depicts an image comparison system of one embodiment of a user interface 500. 用户界面500包括多个已显示的图像510。 User interface 500 includes an image of more than 510 have been shown. 界面500包括针对每个图像的基图像选择按钮(“BASE”)520,其允许将图像指定为基图像以用于比较。 Interface 500 includes a selection button image group for each image ( "BASE") 520, which allows the image to the image specified as a base for comparison. 在一个示例中,一旦选择,基图像选择按钮520例如通过颜色的变化在视觉上加以区分。 In one example, upon selection, for example, the base image selection button 520 to be visually distinguishable by color change. 基图像的位置是基图像显示区域。 Position of the base image is the image display area group. 在选择时,其余图像被指定为分析图像,以便与基图像比较。 When selected, the rest of the image is designated as image analysis, for comparison with the base image. 而且,选择可以触发部分选择窗口的显示。 Moreover, the choice may trigger the display section selection window.

[0062] 分析图像的位置是分析图像显示区域。 [0062] The position of the image analysis is to analyze the image display area. 基图像和分析图像显示区域可以是独立的、离散的显示区域(图5C),也可以不是(图5B)。 Group image display images and the analyzed area may be a separate, discrete display area (FIG. 5C), or may not (FIG. 5B). 如图5A的示例所示,不同的图像对应于在不同的浏览器应用(例如Jnternet Explorer 8,Chrome 4,Firefox 3. 6,Local Safari4,以及LocalOpera 10)中绘制的网页,如每个图像的底部所示。 The example shown in FIG. 5A, the different images corresponding to different pages in a browser application rendering as each image (Jnternet Explorer 8, Chrome 4, Firefox 3. 6, Local Safari4, and e.g. LocalOpera 10) of shown at the bottom.

[0063] 图6描绘了例如弹出窗口形式的部分选择窗口600的一个实施方式。 [0063] FIG 6 depicts an example embodiment of a pop-up selection window 600 forms part of. 窗口600包括用于选择一个或多个部分610的功能,以便按照在此描述的方法在图像间进行比较。 Window 600 comprises means for selecting one or more portions 610 of the function for comparison between the images according to the method described herein. 部分包括各种大小、形状和选择方式。 Moieties include various sizes, shapes, and mode selection. 一旦选择了所有部分,便包括用于发起图像间比较的功能(例如,按钮620)。 Once all parts of the selection is made, including for initiating an inter-image comparison function (eg, button 620).

[0064] 图5B描绘了分析已经开始时图5A的用户界面500的一个实施方式。 [0064] Figure 5B depicts a user interface at the beginning of analysis has an embodiment of FIG. 5A 500. 注意,已经针对图像510a选择了基图像选择按钮520,由此将图像510a指定为基图像,并将图像510b-510e指定为分析图像。 Note that the group has been selected image selection button 520 for the image 510a, whereby the image as the base image designated 510a, 510b-510e and the image is specified as the image analysis. 进度条530示出了分析的进度。 The progress bar 530 shows the progress of the analysis.

[0065] 图5C示出了分析已经完成时图5A的用户界面500的一个实施方式。 [0065] FIG 5C illustrates an embodiment of a user interface 500 of FIG. 5A analysis has been completed. 用户界面500已被分为三个视觉上有区别的并且同时显示的区域:基图像显示区域讨0 (显示选择520以进行比较的基图像510a),分析图像显示区域550(显示分析图像510b_510d),以及相似性范围显示区域560 (指示显示所有图像,也即相似性得分为0或更多的图像)。 The user interface 500 has been divided into three distinct regions and to display visually: 0 discussion group image display region (display selection group 520 for image comparison 510a), analyzes the image display region 550 (image display analysis 510b_510d) and similarity between region display range 560 (indicated show all images, i.e., a similarity score is 0 or more images). 每个分析图像510b-510d现在显示相关联的相似性得分570。 Each image is analyzed 510b-510d show a similarity score 570 is now associated.

[0066] 图5D描绘了应用筛选时图5A的用户界面500的一个实施方式。 [0066] Figure 5D depicts a user interface of one embodiment 500 of the embodiment of FIG. 5A screening application. 在此描绘中,相似性范围显示区域560包括滑块580,用于调整用于分析图像显示区域550中的分析图像显示的阈值相似性。 In this drawing, the similarity range of the display region 560 includes a slider 580 for adjusting the image analysis for analyzing the image display area 550 displays a similarity threshold. 在此示例中,滑块580被设为88,并且由此调整显示的分析图像;只显示了分析图像510b,其相似性得分570为88。 In this example, the slider 580 is set to 88, and thereby adjust the displayed image is analyzed; analysis shows only the image 510b, which is a similarity score 570 88.

[0067] 图5E描绘了显示多个相似性级别时图5A的用户界面500的一个实施方式。 [0067] Figure 5E illustrates one embodiment of a user interface 500 of FIG. 5A displaying the plurality of similarity levels. 在此描绘中,相似性范围显示区域560包括滑块590,其类似于图5D的滑块580,区别在于其显式了各种相似性级别。 In this drawing, the similarity range of the display region 560 includes a slider 590, which is similar to slider 580 5D, with the difference that it explicitly various similarity level. 在所绘示例中,低于30的相似性得分与30到70之间的得分以及高于70的相似性得分在视觉上加以区分。 In the depicted example, the score is lower than the similarity score between 30 and 30 and 70 and above the similarity score is 70 to distinguish visually. 图例区域595说明了每个视觉上区分的级别属于什么。 Legend area 595 illustrates the distinction between what belongs to each visual level. 在此示例中,级别使用实心条、斜线条和点状条显示,然而可以使用视觉上区分级别的任何方式,例如使用各种颜色。 In this example, the level of use of solid bars, oblique lines, and dot bar display, however, in any way visually distinguishing level may be used, for example, using a variety of colors. 图例区域595中显示的较高阈值和较低阈值的每一个可以独立地设置。 Legend display region 595 can be set independently for each higher threshold and lower threshold. 而且,分析图像510b-510d的相似性得分570按照与滑块级别(例如,点570a和斜线570b)相同的视觉区分方式进行显示。 Furthermore, analysis of the image 510b-510d similarity score 570 is displayed in accordance with the slide level (e.g., points 570a and 570b shaded) to distinguish visually the same manner.

[0068] 已经特别详细地关于一个可行的实施方式描述了本发明。 [0068] have a particular detail with respect to possible embodiments of the present invention is described. 本领域技术人员将会理解,可以在其他实施方式中实践本发明。 Those skilled in the art will appreciate that the present invention may be practiced in other embodiments. 首先,组件、术语大小写、属性、数据结构或者任何其他编程或者结构化方面的特殊命名并不是强制性或者具有特殊意义,而是,实现本发明或者其特征的机制可以具有不同名称、格式或者协议。 First, specific named components, capitalization terms, the attributes, data structures, or any other programming or structural aspect is not mandatory or special significance, but the mechanism of the present invention is achieved or its features may have different names, formats, or protocol. 此外,系统可以经由硬件或者软件的结合来实现,或者如所述完全以硬件元件实现。 Further, the system may be implemented via a combination of hardware or software, or as implemented entirely in hardware elements. 另外,在此所述的各种系统组件之间的特定功能划分仅在于示例性而并非强制性;由单一系统组件执行的功能可以替代地由多个组件执行,而由多个组件执行的功能可以由单个组件执行。 Further, the particular division of functionality between the various system components described herein is merely exemplary and that not mandatory; functions performed by a single system component may instead be performed by multiple components, and functions performed by multiple components It may be performed by a single component.

[0069] 上述描述的某些部分以针对信息的操作的算法和符号表示的方式呈现了本发明的特征。 Some parts of [0069] the manner described above algorithms and symbolic representations of operations information presentation features of the present invention. 这些算法描述和表示是由数据处理领域技术人员使用的方式,并且该方式向本领域的其他技术人员最有效地传达其工作的实质。 These algorithmic descriptions and representations are the ways used by those skilled in the data processing art, and ways to most effectively convey the substance of their work to others skilled in the art. 尽管以功能或者逻辑方式描述,应当理解这些操作由计算机程序执行。 Although described functionally or logically, it should be understood that these operations are performed by a computer program. 此外,还应当注意,为方便起见,在不丧失通用性的情况下,将这些操作的设置称作模块或者由功能性名称表示。 In addition, it should be noted that, for convenience, without loss of generality, the set of these operations as modules or by functional names represented.

[0070] 应当理解,除非特别指出,从上述讨论中易见的是,贯穿说明书,使用诸如“确定”或者“显示”等术语的讨论是指计算机系统或者类似电子计算设备的动作和处理,该电子计算设备操纵以及将数据转变为在计算机系统存储器或者寄存器或者其他此类信息存储、传输或者显示设备内表示的物理(电子)量。 [0070] It should be understood that, unless otherwise indicated, from the above discussion is easy to see that throughout the description, terms such as "determining" or "displaying" or the like terms refer to a computer system, or similar action and processes of an electronic computing device, that manipulating the electronic computing device and data into the computer system memories or registers or other such information storage device represented as physical (electronic) quantities transmission or display.

[0071] 本发明的特定方面包括在此以算法形式描述的处理步骤和指令。 Particular aspect of the [0071] present invention include process steps and instructions described herein in the form of an algorithm. 应当注意,本发明的处理步骤和指令可以以软件、固件或者硬件方式实现,以及当以软件实现时,可以将其下载以便驻留在或者从实时网络操作系统使用的不同平台来操作。 It should be noted that the process steps and instructions of the present invention may be implemented in software, firmware or hardware, and when embodied in software, could be downloaded to reside on different platforms from a real-time operating system for use in a network or to operate.

[0072] 本发明还涉及执行在此所述操作的装置。 [0072] The present invention further relates to apparatus for performing the operations herein. 此装置可以是针对所需目的专门构造,或者可以包括由存储于计算机可读介质上的计算机程序来选择激活或者重配置的通用计算机,而该计算机可读介质可以由该计算机访问。 This apparatus may be specially constructed for the required purposes, or may comprise a computer program stored on a computer-readable medium activated or reconfigured to select a general purpose computer, and the computer readable medium may be accessed by the computer. 此类计算机程序可以存储在有形的计算机可读存储介质中,诸如但不限于任何类型的盘,包括软盘、光盘、CD-ROM、磁光盘、只读存储器(ROM)、随机访问存储器(RAM)、EPR0M、EEPR0M、磁性或者光学卡、专用集成电路(ASIC)或者适于存储电子指令的任何类型的介质,以及其中的每个耦合至计算机系统总线。 Such a computer program may be stored on a tangible computer-readable storage medium, such as, but not limited to, any type of disk including floppy disks, optical disks, CD-ROM, magneto-optical disk, read only memory (ROM), a random access memory (RAM) , EPR0M, EEPR0M, magnetic or optical cards, any type of media application specific integrated circuit (ASIC) or suitable for storing electronic instructions, and each of which is coupled to a computer system bus. 此外,在说明书中所指的计算机可以包括单处理器架构,或者可以是为增强计算能力而使用多个处理器设计的架构。 Further, the computer referred to in the specification may include a single processor architecture, or may be used to enhance the ability of computing a plurality of processor design architecture.

[0073] 在此给出的算法和操作并不固有地关联于任何特定计算机或者其他装置。 [0073] The algorithms and operations presented herein are not inherently related to any particular computer or other apparatus. 还可以使用具有根据在此给出教导的程序的各种通用系统,或者还可以证明易于构造更为专用的装置来执行所需的方法步骤。 It can also be used with various general-purpose systems according to a program given the teachings herein, or it may also prove convenient to construct more specialized apparatus to perform the required method steps readily to. 用于各种此类系统所需的结构及其等效变形对于本领域技术人员是易见的。 Such systems for various desired structural modification and equivalents of ordinary skill in the art is readily apparent. 另外,并不参考任何具体编程语言来描述本发明。 Further, no reference to any particular programming language of the present invention will be described. 应当理解,可以使用各种编程语言来执行在此所述的本发明的教导,并且对具体语言的任何参考是提供用于公开本发明的最佳模式和支持。 It should be understood that a variety of programming languages ​​may be used to implement the teachings of the present invention described herein, and any references to specific languages ​​are provided for disclosure and support the best mode of the present invention.

[0074] 本发明同样适于基于多种拓扑的广泛类型的计算机网络系统。 [0074] The present invention is equally applicable to a wide variety of types of computer network topologies based system. 在此领域中,大的网络的配置和管理包括存储设备和计算机,其通过网络(诸如因特网)可通信地耦合至相异的计算机和存储设备。 In this field, the configuration and large networks include storage devices and management computer which through a network (such as the Internet) may be communicatively coupled to dissimilar computers and storage devices.

[0075] 最后,应当注意,在此说明书中使用的语言主要是出于可读性和指示性目的而选择,以及还可以选择用于描绘或者限制创造性的主题。 [0075] Finally, it should be noted that the language used in this specification primarily for readability and instructional purposes, selection, and may also be selected to delineate or circumscribe the inventive subject matter. 由此,本发明的公开旨在于示意性的,而并非限制本发明的范围,本发明的范围由所附权利要求书来限定。 Accordingly, the disclosure of the present invention is intended to be illustrative, and not to limit the scope of the present invention, the scope of the invention defined by the appended claims.

13 13

Claims (29)

1. 一种用于比较图像间相似性的计算机实现的方法,包括:在第一计算机处,接收基图像和所述基图像的选定部分,作为图像比较的基础;确定所述选定部分的颜色信息;在所述第一计算机处,接收至少一个分析图像;确定所述至少一个分析图像与所述基图像的所述选定部分的对应部分;确定所述至少一个分析图像的所述对应部分的颜色信息;将所述基图像的所述选定部分的所述颜色信息与所述至少一个分析图像的所述对应部分的所述颜色信息进行比较;以及在所述第一计算机处,基于所述比较来确定所述基图像与所述至少一个分析图像之间的相似性。 1. A method for comparing image similarity between computer-implemented, comprising: a selected portion of the first computer, and said image receiving substrate base picture, as a basis for comparison of the image; determining the selected portion color information; at the first computer, receiving at least one image analysis; determining at least one analysis of said portion of said image corresponding to the selected base image; determining the at least one image analysis color information corresponding portion; the color of the color information of the corresponding portion of the base image with the selected portion of the at least one image analysis comparing the information; and said first computer , determining the base image and the at least one analysis of the similarity between the images based on the comparison.
2.如权利要求1所述的方法,其中所述基图像和所述至少一个分析图像对应于由不同设备绘制的网页。 2. The method according to claim 1, wherein the base image and the image corresponding to the at least one analyte page drawn by different devices.
3.如权利要求1所述的方法,还包括:接收与所述选定部分的选择相对应的坐标。 The method according to claim 1, further comprising: receiving the selected portion corresponding to the selected coordinates.
4.如权利要求1所述的方法,其中所述基图像是网页,并且所述选定部分是整个所述网页。 4. The method according to claim 1, wherein the substrate is a web image, and the selected portion of the entire web.
5.如权利要求1所述的方法,其中所述基图像是网页,并且所述选定部分排除所述网页的区块。 5. The method according to claim 1, wherein the substrate is a web image, and the selected portion of the negative block page.
6.如权利要求1所述的方法,其中接收的所述至少一个分析图像包括多个分析图像,并且在所述基图像与所述多个分析图像的每一个之间比较所述颜色信息,以确定所述基图像与所述多个分析图像的每一个之间的相似性。 6. The method according to claim 1, wherein the at least one received image analysis comprises analyzing a plurality of images, and analyzing the comparison between each of said color image information in the image and the plurality of groups, to determine the base image and the plurality of analysis of the similarity between each image.
7.如权利要求1所述的方法,其中确定颜色信息还包括:量化所述基图像的所述选定部分中的颜色;以及量化所述至少一个分析图像的所述对应部分中的颜色;其中比较颜色信息还包括:将所述基图像的所述选定部分中的量化颜色与所述至少一个分析图像的所述对应部分中的量化颜色进行比较;以及其中确定所述基图像与所述至少一个分析图像之间的相似性包括:基于所述基图像的所述选定部分中的所述量化颜色与所述至少一个分析图像的所述对应部分中的所述量化颜色之间的相似性,确定相似性得分。 7. The method according to claim 1, wherein the color determining information further comprises: quantizing the image of the base color of the selected portion; and at least one analysis of said quantizing portions corresponding to the color of the image; wherein comparing the color information further comprises: quantizing the color image of the selected base portion and said at least one analyte quantizing the color image in the corresponding portion of the comparing; and wherein determining the base image and the said at least one analyte similarity between images comprising: a base image based on said selected portion of the quantization of the color of the image between the at least a corresponding portion of the quantized color analysis similarity, a similarity score is determined.
8.如权利要求7所述的方法,其中量化所述选定部分以及所述对应部分中的颜色包括:确定所述选定部分以及所述对应部分中的颜色的像素数目。 8. The method according to claim 7, wherein the quantizing portion and the selected portion corresponding to a color comprising: determining a number of the pixel portion and the corresponding portion of the selected color.
9.如权利要求7所述的方法,其中确定所述相似性得分还包括:基于所述基图像的所述选定部分中的所述量化颜色与所述至少一个分析图像的所述对应部分中的所述量化颜色之间的相似性,确定部分的部分得分;以及将所述部分得分除以所述选定部分的总大小。 9. The method according to claim 7, wherein determining the similarity score further comprises: quantizing the color image corresponding to at least a portion of the base image analysis based on the selected portion the quantization of similarity between the color determining portion scores; and the score portion divided by the total size of the selected portion.
10.如权利要求9所述的方法,其中确定所述部分得分还包括:确定颜色的颜色得分;以及将所述颜色得分与所述部分的其他颜色得分相结合,以产生所述部分得分。 10. The method according to claim 9, wherein determining said score portion further comprising: determining the color of the color scores; and the other colors of the color scores combined score portion, the portion to generate a score.
11.如权利要求1所述的方法,其中确定所述颜色信息还包括:量化所述基图像的所述选定部分中的颜色;以及量化所述至少一个分析图像的所述对应部分中的颜色;其中比较所述颜色信息还包括:进行对所述基图像的所述选定部分中的量化颜色的第一散列;以及进行对所述至少一个分析图像的所述对应部分中的量化颜色的第二散列;以及其中确定所述基图像与所述至少一个分析图像之间的相似性还包括:比较所述第一散列与所述第二散列,以确定所述基图像的所述选定部分与所述至少一个分析图像的所述对应部分中的量化颜色是否等同;响应于确定量化颜色等同,指派最高相似性得分;以及响应于确定量化颜色不等同,确定量的相似性得分。 11. The method according to claim 1, wherein said color determining information further comprises: quantizing the image of the base color of the selected portion; and quantizing the at least one analysis of said image corresponding portions color; color information wherein said comparing further comprises: a first hash quantizing color of a selected portion of the image group; and quantizing the corresponding portion of said at least one analyte in the image second hash color; and wherein the determining the at least one image based analysis of the similarity between the image further comprises: comparing the first image of the group hash and the second hash to determine the selected portion of the at least one corresponding portion of the image analysis of the color is equal quantization; in response to determining that the quantized color equivalents, assigned the highest similarity score; and in response to determining that the quantized colors are not identical, determining the amount of similarity score.
12.如权利要求1所述的方法,还包括:存储所述基图像与所述至少一个分析图像之间的相似性。 12. The method according to claim 1, further comprising: storing the base image and the at least one analysis of the similarity between the images.
13.如权利要求1所述的方法,还包括:基于作为与所述至少一个分析图像的每一个相关联的相似性得分的比较,显示所述基图像与所述至少一个分析图像之间的相似性。 13. The method according to claim 1, further comprising: based on a comparison with the at least one similarity score associated with each of the image analysis, display the image with at least one group between the image analysis similarity.
14.如权利要求13所述的方法,还包括:显示用于选择阈值相似性的控件;以及更新显示以示出所述至少一个分析图像中得分大于所述阈值相似性的每一个的子集。 14. The method according to claim 13, further comprising: displaying a selection threshold similarity controls; and update the display to show at least one image analysis score greater than the threshold value for each subset of similarity .
15.如权利要求14所述的方法,其中所述控件选择较低阈值相似性,还包括:显示第二控件,以选择较高阈值相似性;其中更新所述显示包括:提供所述较高阈值相似性之上、所述较低阈值相似性之下、以及所述较低阈值相似性和较高阈值相似性之间的得分的可视指示。 15. The method according to claim 14, wherein the control selects a lower threshold of similarity, further comprising: a second display control to select a higher threshold of similarity; wherein updating the display comprising: providing the higher a similarity above a threshold, below the lower threshold of similarity, and the lower threshold value and upper threshold value of similarity scores between visual indication of similarity.
16.如权利要求1所述的方法,其中基于所述比较来确定所述基图像与所述至少一个分析图像之间的相似性还包括:确定表示所述基图像的所述选定部分的所述颜色信息与所述至少一个分析图像的所述对应部分的所述颜色信息之间差异的差异得分;以及通过向最低的差异得分指派最高的相似性,来建立相似性。 16. The method according to claim 1, wherein determining based on the comparison of the base image and the at least one analysis of the similarity between the image further comprises: determining a group represented by the selected image portion the color information analyzing the at least one difference between the color difference corresponding to the portion of the image information of the score; and assigning the highest similarity score to the lowest difference is established by a similarity.
17. 一种有形计算机可读存储介质,具有包含于其中的计算机程序指令,用于比较图像间的相似性,包括:图像模块,配置用于接收基图像和所述基图像的选定部分以作为图像比较的基础,以及接收至少一个分析图像,并确定所述至少一个分析图像与所述基图像的所述选定部分的对应部分;颜色信息模块,配置用于确定所述选定部分的颜色信息以及所述至少一个分析图像的所述对应部分的颜色信息;比较模块,配置用于将所述基图像的所述选定部分的所述颜色信息与所述至少一个分析图像的所述对应部分的所述颜色信息进行比较,以及基于所述比较来确定所述基图像与所述至少一个分析图像之间的相似性。 17. A tangible computer readable storage medium having computer program instructions contained therein, for comparing the similarity between images, comprising: receiving a selected portion of the base image and the image in the image group module configured to as a basis for comparison of the images, and receiving at least one image analysis, and determining said at least one analysis corresponding to the portion of the base image and the selected image; color information module, configured to determine the selected portion color information and color information of the at least one analysis corresponding to the portion of the image; comparing module, configured for the color information of the image group to the selected portion of the at least one image analysis the color information corresponding to the comparing portion, and wherein said base determined by comparing the at least one image analysis of the similarity between images based.
18.如权利要求17所述的有形计算机可读存储介质,其中所述基图像和所述至少一个分析图像对应于由不同设备绘制的网页。 18. wherein the base image and the image corresponding to the at least one analyte page drawn by different apparatus as claimed in claim 17 tangible computer-readable storage medium.
19.如权利要求17所述的有形计算机可读存储介质,其中所述基图像是网页,并且所述选定部分排除所述网页的区块。 19. The tangible computer readable storage medium of claim 17, wherein the substrate is a web image, and the selected portion of the negative block page.
20.如权利要求17所述的有形计算机可读存储介质,其中所述颜色信息模块还配置用于:确定所述基图像的所述选定部分中的颜色的像素数目,以及确定所述至少一个分析图像的所述对应部分中的颜色的像素数目;其中所述比较模块还配置用于:对所述基图像的所述选定部分中的颜色的像素数目与所述至少一个分析图像的所述对应部分中的颜色的像素数目进行比较;以及进一步包括计分模块,配置用于:基于所述基图像的所述选定部分中的颜色的像素数目与所述至少一个分析图像的所述对应部分中的颜色的像素数目之间的相似性,确定相似性得分。 20. The tangible computer readable storage medium according to claim 17, wherein the color information module is further configured to: determine the number of pixels of the portion of the base color of the selected image, and determining at least analysis of a portion of the number of pixels of the image corresponding to a color; wherein the comparison module is further configured to: group the image of the selected portion of the number of pixels of the at least one color image analysis the number of pixels in the corresponding portion of the color comparing; and further comprising a scoring module configured to: analyze at least one image based on the number of pixels of the base image with the selected color portion said number corresponding to the similarity between the color of the pixel portion, and determining a similarity score.
21. 一种用于比较基图像的选定部分与至少一个分析图像的对应部分之间的相似性的系统,包括:用于确定所述选定部分的颜色信息以及所述至少一个分析图像的所述对应部分的颜色信息的装置;用于将所述基图像的所述选定部分的颜色信息与所述至少一个分析图像的所述对应部分的颜色信息进行比较的装置;以及用于基于所述比较来确定所述基图像与所述至少一个分析图像之间的相似性得分的装置。 21. A method for comparing a selected portion of the image with at least one group analysis of the similarity between the corresponding portions of the image system, comprising: means for determining the color information of the selected portion and the at least one image analysis means the color information of the corresponding section; base image for the color information of the selected portion of the at least one analyte color information corresponding portion of the means for comparing the image; and based on said comparison determining the base image and said at least one similarity score between the image analysis.
22.如权利要求21所述的系统,其中所述基图像和所述至少一个分析图像对应于由不同设备绘制的网页。 22. wherein the base image and the image corresponding to the at least one analyte page drawn by different apparatus system as claimed in claim 21,.
23.如权利要求21所述的系统,其中所述基图像是网页,并且所述选定部分排除所述网页的区块。 23. The system according to claim 21, wherein the substrate is a web image, and the selected portion of the negative block page.
24.如权利要求21所述的系统,其中所述选定部分的颜色信息包括所述基图像中的所述选定部分中的颜色的像素数目,并且所述对应部分的颜色信息包括所述至少一个分析图像的所述对应部分中的颜色的像素数目;以及还包括:用于基于所述基图像的所述选定部分中的颜色的像素数目与所述至少一个分析图像的所述对应部分中的颜色的像素数目之间的相似性来确定相似性得分的装置。 24. The system according to claim 21, wherein said selected portion comprises a number of color information of the pixel portion of the color image in the selected group, and the color information corresponding to said portion comprises at least a partial image corresponding to the number of pixels in the color analysis; and further comprising: a base image based on said selected number of colors in the pixel portion with the at least one analysis of said image corresponding means for determining a similarity between the similarity score of the number of pixels of the color portion.
25. 一种有形计算机可读介质,存储处理器可执行的计算机程序,所述计算机程序产生图像比较系统的用户接口,所述用户接口包括:基图像显示区域,用于显示基图像;分析图像显示区域,其与所述基图像显示在区域视觉上有区别地同时显示,用于显示至少一个分析图像以及与所述至少一个分析图像的每一个相关联的得分;以及相似性范围显示区域,其与所述基图像显示区域和所述分析图像显示区域在视觉上有区别地同时显示,用于显示对应于所述基图像与所述至少一个分析图像之间的相似性的相似性信息。 25. A tangible computer readable medium storing a computer program executable by a processor, the computer program to generate an image comparison system user interface, the user interface comprising: a base image display area for displaying an image group; Image Analysis display region, with said base there are differences in the image display area while visually display for displaying images and at least one analysis of said at least one score associated with each of the image analysis; and similar range of the display region, with said base region and said analysis image display image display area visually discriminately display at the same time, for displaying an image corresponding to said group with at least one similarity information of the similarity between the image analysis.
26.如权利要求25所述的有形计算机可读介质,其中所述相似性范围显示区域还包括:相似性标尺,显示用于选择所述至少一个分析图像的显示的阈值相似性的控件。 26. The tangible computer-readable medium of claim 25, wherein the similarity range display area further comprises: a similarity scale, for selecting at least one threshold value of similarity of the image display control analysis.
27.如权利要求沈所述的有形计算机可读介质,还包括:响应于调整用于选择所述阈值相似性的所述控件的输入,调整所显示的所述至少一个分析图像。 27. The tangible computer readable medium of claim Shen, further comprising: in response to the adjustment input for selecting the threshold value of the similarity control adjusting at least one of said displayed image analysis.
28.如权利要求沈所述的有形计算机可读介质,其中所述相似性标尺显示所述基图像与所述至少一个分析图像之间的相似性级别。 28. The tangible computer readable medium of claim sink, wherein said display scale of the similarity groups with the at least one image analysis of the similarity between the level of the image.
29.如权利要求25所述的有形计算机可读介质,还包括:用于指定所述基图像以及用于选择所述基图像的至少一部分以便与所述至少一个分析图像进行比较的控件。 29. The tangible computer-readable medium of claim 25, further comprising: means for specifying the base picture of the base image and means for selecting at least one control for image analysis and comparing said at least a portion.
CN201010260502XA 2010-08-24 2010-08-24 System for quantizing similarity between images on computer CN102376088A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201010260502XA CN102376088A (en) 2010-08-24 2010-08-24 System for quantizing similarity between images on computer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201010260502XA CN102376088A (en) 2010-08-24 2010-08-24 System for quantizing similarity between images on computer

Publications (1)

Publication Number Publication Date
CN102376088A true CN102376088A (en) 2012-03-14

Family

ID=45794641

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201010260502XA CN102376088A (en) 2010-08-24 2010-08-24 System for quantizing similarity between images on computer

Country Status (1)

Country Link
CN (1) CN102376088A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103810098A (en) * 2012-11-04 2014-05-21 正谓有限公司 Evaluation of resizing capability of web browser
CN104111960A (en) * 2013-04-22 2014-10-22 阿里巴巴集团控股有限公司 Page matching method and device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1293783A (en) * 1999-02-01 2001-05-02 Lg电子株式会社 Muctilevel image grid data structure and image search method using the same
US6999636B1 (en) * 1999-11-09 2006-02-14 Canon Kabushiki Kaisha Image search method and apparatus
CN1926575A (en) * 2004-03-03 2007-03-07 日本电气株式会社 Image similarity calculation system, image search system, image similarity calculation method, and image similarity calculation program
CN101300575A (en) * 2005-10-31 2008-11-05 索尼英国有限公司 Image Processing

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1293783A (en) * 1999-02-01 2001-05-02 Lg电子株式会社 Muctilevel image grid data structure and image search method using the same
US6999636B1 (en) * 1999-11-09 2006-02-14 Canon Kabushiki Kaisha Image search method and apparatus
CN1926575A (en) * 2004-03-03 2007-03-07 日本电气株式会社 Image similarity calculation system, image search system, image similarity calculation method, and image similarity calculation program
CN101300575A (en) * 2005-10-31 2008-11-05 索尼英国有限公司 Image Processing

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103810098A (en) * 2012-11-04 2014-05-21 正谓有限公司 Evaluation of resizing capability of web browser
CN104111960A (en) * 2013-04-22 2014-10-22 阿里巴巴集团控股有限公司 Page matching method and device

Similar Documents

Publication Publication Date Title
JP5237469B2 (en) Display multiple row and column header areas in summary tables
Verdoolaege isl: An integer set library for the polyhedral model
US8326091B1 (en) Ranking of images and image labels
CN101573705B (en) Media material analysis of continuing article portions
US9418319B2 (en) Object detection using cascaded convolutional neural networks
CN100476827C (en) Information processing apparatus and information processing method
US8811734B2 (en) Color determination device, color determination system, color determination method, information recording medium, and program
US8560940B2 (en) Detecting repeat patterns on a web page using signals
CN102105901B (en) Annotating images
US8515208B2 (en) Method for document to template alignment
Bae et al. High-precision vision-based mobile augmented reality system for context-aware architectural, engineering, construction and facility management (AEC/FM) applications
JP2007164648A (en) Similar image search device, similar image search method, program and information recording medium
US8891860B2 (en) Color name determination device, color name determination method, information recording medium, and program
US20070065045A1 (en) Information management apparatus, information management method, and computer program product
Shihab et al. Understanding the impact of code and process metrics on post-release defects: a case study on the eclipse project
US20120102388A1 (en) Text segmentation of a document
CN103080924B (en) Method and apparatus for processing data set
CN101627399B (en) Feature matching method
US8977520B2 (en) Computer system for automatically classifying roof elements
CN103329126A (en) Search with joint image-audio queries
US20130272627A1 (en) Methods and systems for processing a first image with reference to a second image
TWI654567B (en) Method and apparatus for extracting specific information from standard cards
CN100414549C (en) Image search system, image search method, and storage medium
US20110078176A1 (en) Image search apparatus and method
CN102236693B (en) Method and device for determining similarity between documents

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)