MXPA00007868A - Network image view server using efficient client-server, tiling and caching architecture - Google Patents

Abstract

A computer network server using HTTP (Web) server software combined with foreground view composer software (50), background view composer software (80), view tile cache (60), view tile cache garbage collector (70) and image files (90) provides image view data to client workstations (20) using graphical web browsers to display the view of an image from the server. Problems with specialized client workstation image view software are eliminated by using the internet and industry standards based graphical web browsers for the client software. Network and system performance problems that previously existed when accessing large image files from a network file server are eliminated by tiling the image view so that computation and transmission of the view data can be done in an incremental fashion. The vied tiles are cached on the client workstation to further reduce network traffic. View tiles are cached on the server to reduce the amount to view tile computation and to increase responsiveness of the image view server.

Description

NETWORK IMAGE VISION SERVICE USING EFFICIENT CUSTOMER-SERVER, PACKAGED AND CACHE ARCHITECTURE BACKGROUND OF THE INVENTION 1. Field of the Invention This invention relates to a workstation for viewing images of digital documents stored on a network server, and in particular, viewing large images of digital documents using a client-server architecture. 2. Description of the Prior Art Current methods for viewing images of digital documents for workstations in a networked environment use proprietary workstation application software to access a network image file server. To view an image, the application software transfers a copy of the entire image file from the image file server to the network client workstation. This method has a number of limitations that include: inefficient use of the network; high cost of software purchase per workstation; high administrative cost of the software per work station; high computational demands on the work station; proprietary software available only for limited workstation types. Some other network image views can provide the view using more optimized image transmission protocols but only with proprietary protocol and proprietary workstation software. It is an object of the invention to provide a method for obtaining graphic images from a network server to view on a computer workstation that does not require proprietary workstation software. It is another object to provide such a method that makes efficient the use of the network and of how a larger space of image representation results in response to requests by the work stations. It is another object to provide a method that uses cache mechanisms resulting in a balanced load on the network file server and a faster response time to a single client when many clients are accessing the server simultaneously. It is another object to minimize the computing resources required by a client of the workstation. A further object is to provide an apparatus for storing graphic images, which require portions of the stored graphic images from storage, and quickly and efficiently represent the images on a workstation. A further object is to provide a computer program that facilitates the request of portions of graphic images stored in a network server and the representation of those portions in a workstation.

BRIEF DESCRIPTION OF THE INVENTION These objects and others, which will be apparent from the following description, are achieved through this invention which comprises in one aspect a method for identifying and supplying a graphic image from a computer network file server comprising providing a network file server where image files of digital documents are stored, said server adapted to receive requests from a Web browser in a Uniform Resource Locator code (URL), to identify the image file and format selections that they are being requested, to compose the vision requested in a grid of tiling of vision, and to transmit an HTML code to see the tiling in the web browser that it requests.

Another aspect of the invention comprises an apparatus comprising a computer network server adapted to store image files of digital documents, programmed to receive requests from a client Web browser in the URL code, the URL specifying a view identifying a file of image and format, to compose the requested view, and to transmit the HTML code for the resulting view to represent it to the client Web browser. A further aspect of the invention is a computer program recorded on a magnetic or optical medium for use on a network server, comprising a code that interprets HTTP requests from a workstation for a particular view of an image file of digital documents stored in memory, retrieve the image file from digital documents, form a grid of tiling of vision that correspond to the requested view of the image, calculate the HTML code for the grid of tiling of vision in a form that can be transmitted from the server to the workstation. The accompanying drawings, which are incorporated and constitute a part of the specification, illustrate one embodiment of the invention and together with the general description serve to explain the principles of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a diagram of the architecture of the system showing the relationship of the components of the system and the image viewing server. Figure 2 is a flow diagram of the steps taken by the system to request, form and represent a view of an image. Figures 3A and 3B are diagrams showing the vision tiling grid as determined by the vision scale. Figures 4A and 4B are diagrams showing the grid vision tiling formed for an initial image view and then for a displaced view of the image. Figures 5A and 5B are diagrams showing the web browser representation of the view tiling for an initial view and then for a displaced view of the image corresponding to Figures 4A and 4B. Figures 6A and 6B are diagrams showing precomputed viewing tiling through the background view former. Figure 7 is a high-level flow diagram of the anterior vision former. Figure 8 is a flow chart for the vision generator component of the vision former. Figure 9 is a flow diagram for the data output component of the vision former. Figures 10A, 10B and 10C together constitute a flowchart for viewing the garbage collector by caching the vision tiling.

DETAILED DESCRIPTION OF THE INVENTION AND THE PREFERRED MODALITIES Reference will now be made in detail to the currently preferred embodiment of the invention, an example of which is illustrated in the accompanying drawings. The preferred mode on a PC server consisting of an Intel Pentium Pro 2000 MHz processor, with at least 128 MB of RAM, an Ultra-wide Fast SCSI disk controller with at least 4 GB of hard disk space, and controllers LAN / WAN / Internet interface. The server runs the operating system version 4 of the Windows NT server with the NT file system, version 3 of the Microsoft Internet information server, and the network image server software. The server and the client are configured with TCP / IP network protocols to support the HTTP (Web) protocol. No software other than a Web browser is required on the client. The preferred Web browser is Internet Explorer 3.0 or Netscape 3.0 or higher. Referring first to Figure 1, a network comprising client workstations 10 and 20 are connected through network connections to a network image viewer 100 comprising a network server interface, preferably a server Web 30 using the Hypertext Transfer Protocol (HTTP), a request broker 40, a previous viewer 50, a vision tiling cache 60, a background vision former 80, a garbage collector 70 and a Document repository 90 having image files. The network image viewing server, that is, the client workstation, or "workstation" 100 can be implemented in a computer, for example, a personal computer configured with a processor I / O memory, storage in disk, and a network interface. The network image view server 100 is configured with a network server operating system and Web server software 30 to provide the network HTTP protocol link with the client workstations 10 and 20. Typical networks include: many workstations served by one, and sometimes by more than one network server, the server functioning as a library to maintain files that can be accessed by workstations. During operation, according to one embodiment of the method of the invention, using the Web browser software on the client workstation, a user requests an image view 110 (Figure 2) having a scale and regions specified by means of a uniformly formatted resource locator code (URL) using the HTTP language, which the Web server can decode as a request that will be passed to an image viewing composition software and that identifies the image file to be viewed, the scale of the view and the region of the image to be viewed. The network image server sends the HTML data to the client with precomputer hyperlinks, so that following a hyperlink by clicking on an area of an image will send a specific request to the server to supply a different area of the drawing or change the resolution of the image. The HTML resulting from this request will also contain precomputed hyperlinks for other options that the user can exercise. The code is sent over the network to the network server, where the Web server software interprets request 120, passes the request request URL to the previous viewer training software through a common composite interface (CGI) ) which is designed to allow the processing of HTTP requests outside the Web server software, and thus instructs the request broker 130 to obtain the particular required view, having the scale and region requested by URL. The former viewer is initialized to 140 and forms the requested view 150 after retrieving it from the memory in the network server. The former viewer software interprets the vision request, calculates which vision tiles are necessary for the view, creates the view tiling 160 needed for the view, and then creates the output file for Hypertext Markup Language (HTML) ) to describe the composition of the view to the Web browser, unless the tiling of vision needed to fill the request is already calculated and stored in the workstation's cache memory, in that case, the already computed tiles are retrieved by the Web browser. In any case, the former view former formats the output 170 and then initializes the background view former 180 which passes the formatted output to the Web server, which in turn transmits the formatted output over the network to the Web browser 200 in the request workstation 10, wherein the requesting examiner deploys any already cached vision tiling 210, combined with the newly computer vision tiling 220, which are brought from the server. The generation of tiling of vision 160 is handled by an image tiling routine, which divides a given page, the shape as an image, into a grid of smaller images (Figure 3A) called viewing tiling A1, A2, B1 , etc. (or just tiling in the context of the image view server). These tiling are computed for different resolutions (Figure 3) of a given image on the server according to the URL request received from the examiner software on the workstation. The use of tiling allows a cache of effective image data 60 in the image viewing server and by the examiner 10 in the client workstation. The preferred view tiling format is for 128-pixel or 128-pixel GIF image files. The GIF image file format is preferred, due to the compatibility of the Web browser and the image file size. The GIF image file format is the most widely supported format for graphical Web browsers and therefore gives maximum client compatibility for the image viewing server. The GIF image format has the desired compression properties of loss image data, reasonable data compression ratios, color and gray scale support, and a relatively small image file header, which refers to the selection of the size of vision tiling. With a starting image data size for monochromatic 2048-bit tiling and a typical GIF compression of 4 to 1, the compressed data for a tiling of vision is approximately 512 bits. With many image file formats, such as TIFF and JPEG, the image file head (and other superior information such as data index) may be as large or larger than the same image data for small images such as tiling of vision; while a GIF header for a monochrome image adds as little as 31 bits to the GIF image file. Alternative vision tiling formats such as Portable Network Graphics (PNG) can be used, especially as a native examiner support for the format that becomes common. The 128-pixel viewing tiling size is a good compromise between the granularity of tiling of vision and the top of tiling of vision. The 128-pixel tiling view granularity determines the minimum viewing distance (panning distance) that can be obtained with the standard graphic Web browser and HTML level 2 formatting. This allows adjustment of the view position in a 1.62 cm grid when you see an image of 200 pixels by 2.54 cm on a scale of 1 to 1. The reduction of the size of the tiling of vision allows a thinner grid for the placement of the view, but presents the problem that the top part of the tiling of vision becomes excess. A tiling of vision typically represents more or less 128 x 128 pixels of the image file. If the view that is being displayed is reduced from 2 to 1, then each viewing tiling will represent an area of 256 x 256 pixels of the image file that has been scaled to 128 x 128 pixels. For each possible scale factor, there is a tiling arrangement to represent the view. The tiling of vision of size in pixels is beneficial, since it allows a more effective use of the cache mechanism in the server and in the client. For example, consider a view of 512 pixels x 512 pixels. Without tiling, this view is composed of an individual GIF file that is displayed by the Web browser, and so if the user requests that the view be displaced by 156 pixels, then it is necessary to create a new GIF image of 512 x 512 pixels and be transmitted to the Web browser. With tiling, the first view can cause 16 tiling of vision to be calculator and transmitted to be displayed by the Web browser. When the view request is made to be displaced by 256 pixels, only 8 vision tiles representing an area of 256 x 512 pixels need to be calculated. In addition, only the 8 new vision tiling needs to be transmitted to the Web browser, since the displaced view will again use 8 tiling of vision that are available from the Web browser cache. The use of tiling cuts the calculation and transmission of data in half, for this example. The use of vision tiling also allows the image viewing server to effectively precompute tiling of vision that may be required by the next vision request. The background vision trainer of the image vision server calculates the vision tiling surrounding the most recent vision request in anticipation of a request for a displaced view. When the displaced view is requested, the anterior vision former can use the precomputed viewing tiling and eliminate the time to calculate new vision tiling for the view. For frequently accessed images, there is a good chance that view tiling for a view may already exist in the vision tiling cache, since the vision tiling cache maintains the most recently accessed vision tiling. Since millions of tiling of vision can be created and ultimately exceeds the storage capacity of the image viewing server, the tiling collector of tiling of vision removes at least the newly accessed vision tiling in the case where achieve maximum storage distribution limits or minimum storage free space limits. The number of tiling of vision needs to make a given vision size increase in inverse proportion to the square of size of tiling of vision. A 64-pixel vision tiling may require 4 times more vision tiling to make this same viewing area, and this is less preferred. The top part of the tiling of vision exists as a quantity of data and as the number of network transactions. The amount of data that comes from the image file head size as a proportion of the total image file size as described above and how the data is needed to make the tiling references of vision in the HTML text file. The top of network transaction is increased with smaller tiling of vision, since each of the tiling of vision requires a network transaction. The increased number of network transactions required with a smaller vision tiling size can encourage the response to make a view.

The HTML output file produced by the previous viewer is passed to the Web server software to be transmitted to the Web browser. The graphic Web browser serves as an image viewer using the HTML output of the image viewing server to form and display the array of view tiling that forms a view of an image. The list of HTML page data, size, position and hyperlink for each viewing tiling will be displayed. The tiling of vision is stored in the GIF image file format that can be displayed by all common graphical Web browsers. The Web browser will retrieve each tiling of vision that will be displayed from a local cache if the tiling of vision is present, otherwise of the image viewing server. The requesting corridor 40 takes the starting request from the network server interface 130 , interprets the request, communicates with the other components of the system and determines what is the appropriate response. It also determines when the response returns. In the preferred mode, the request broker is implemented with the Web Server Common Gateway (CGI) interface. There are options for using Direct Application Program (API) interfaces for the Web server. To support the tiling and caching architecture of many images on the same image viewing server, each tiling of vision should only be identified for reference by the Web browser with a tiling vision URL. This individuality is achieved through the combination of a storage location and appointment of vision tiling. The individuality between images is achieved by having a separate storage subdirectory in the vision tiling cache for each image. The individuality of tiling of vision for each scale of sight is achieved through the file name for each tiling of vision. The name of the tiling of vision preferably is of the following form: V < SCALE > < TILE_NUMBER > .GIF The value < SCALE > is a character strip of 2 formed from the base 36 that encodes the vision scale number as expressed in parts by 256. The value < TILE_NUMBER > is a character strip of 5 formed from base 36 that encodes the number of tiling as determined by the formula: TILE_NUMBER = TILE_ROW + IMAGE_TILE_WIDTH + TILE_COLUMN The TILE_ROW and TILE_COLUMN values start at 0 for this calculation. For example, the second tiling of the first row of a 2: 1 scaled view can be named under the preferred protocol: V3J00001.GIF The entire URL reference for the second tiling of the first row for an image number 22 on the server Image viewing can be: htt p: / h ost na me / view-t i le-ca chepat h / 000022 / V3 J00001.GIF In addition to the position of the tiling of vision and the scale of vision, other attributes vision can be encoded in the location of storage of tiling of vision or in the name of tiling of vision. These attributes are views of rotation angle, mirror view x, mirror views and, inverted view. A vision tiling name with these extra view attributes can be encoded as: V < SCALE > < TILE_NUMBER > < VIEW_ANGLE > .X_MIRROW > < X_MI RROW > < INVERT > .GIF VIEW_ANGLE is the A < ANGLE > .X_MIRROW, Y_MIRROW, and INVERT are encoded by individual characters X, Y, and I respectively. An example is V3J00001A90XYI.GIF The Web server 30 is configured to recognize the specially formatted, uniformly formatted petition resource locators, described above, (URL) that will be handled by the request broker 40 of the image viewing server. This is done through association of request broker 40 with the URL path or with the file name extension of the document. The previous vision former 50 interprets the vision request command 140 to determine which view needs to be formed. The vision request can be absolute by defining the position scale, relative, defining scale and position, as delta to a preview, or implied based on omissions of the system to select the view. The vision computing software routine 150 is illustrated in Figure 7, where the command interpreter 151 takes the vision request and determines 152 which scale vision tiling grid is necessary for the view and which tiling vision within of the grid are necessary for the view 150 (Figure 2), and generates the tiling of vision 153, resulting in the output of the formatted vision 154. The routine of the tiling generator of vision 160 realizes the creation of the tiling of vision according to the preferred steps shown in Figure 8. The vision tiling generator receives information from the view computation as to which tiling of vision is necessary for the view. It has access to records in cache 80 that determine which tiles have already been created and are resident in the cache. If a required tiling of vision is in the cache, then its last access time is updated to prevent the cached garbage collector from removing the tiling of vision. If a required tiling is not in the cache, then the vision tiling generator creates the tiling of vision of the image file 90. The tiling generator uses a software imaging library that supports many file formats of digital documents including monochrome raster images, gray scale raster images, color raster images as well as many rich content-less formats such as the Adobe Portable Document Format (PDF), PostScript, HPGL, etc. When monochromatic image data is obtained, the grayscale scale of image formation collection is used to provide a visually more attractive appearance of the reduced image. For example, a specific vision request may include tiling B2, C2, B3, and C3 (Figure 4A and Figure 5A). If after seeing these tiling, the client decides that the left-hand view is desired, then the server can send the A2 and A3 tiling (Figure 4A and Figure 5B). This assumes that the client retains the other tiles in the cache. If the client does not keep in cache, then the tiles A2, A3, B2, and B3. A formatted output, 170, is created to refer to the tiling of vision needed to display the full view. The formatted output uses HTML to describe the order, position and hyperlink for each viewing tiling that will be displayed. The output formatter applies another optimization by detecting blank view tiling and replacing the hyperlink for the image-specific white tiling with a common white tiling reference. This eliminates the transmission of all, but a white tiling, and eliminates the need to transmit any white tiling once the common white tiling is cached by the Web browser. The anterior view former 50 controls the background view former 80 by stopping it when a previous view is being formed and then starting with information regarding the new view, once the new view has been formed, 180. Preferably, the former Background 80 is programmed to optimize the operation of the system by precomputing, that is, forming, tiling of vision that may be necessary between the requests of view in the server. The tiles that will be precomputed are the tiles that surround the most recent view and with the same scale as the most recent view. The precomputing order of the tiling of vision is a spiral clockwise from the most recent view. The number of precomputed viewing tiling around a view is preferably limited by a system configuration parameter to avoid unnecessary creation of vision tiling that may never be required. Once the maximum number of precomputed vision tiling is achieved for the most recent view scale, precomputing can proceed for alternating view scales. The background view former is preferably programmed to work at a low priority to minimize interference with more critical system tasks. Figure 6A illustrates how the background vision tracker algorithm works. Assuming that for a given view requested by the client, the tiling C3, C4, D3 and D4 are retrieved, after these tiling are administered to the web browser, the background vision trainer routine within the server program creates the tiling around these tiles, starting at E4, forming or calculating said surrounding tiling. As the client continues to look at this page at this scale factor, the server will calculate the tiling of vision that expands outward from the requested tiling in final form. Figure 6B illustrates another request made by a customer, after generating the two tiling rotations. The petition requested the tiling G3, G4, H3, and H4. When tiling precomputing begins this request will create the tiling G5, H5, 15, 14, 13, 12, H2 and G2 in the first rotation but will not attempt to create tiling in column F. Preferably, a cache garbage collector algorithm 70 tiling tiling handles the use of storage for vision tiling (Figures 10A, 10B, 10C). The garbage collector maintains the cache 60 of the tiling of vision (Figure 1) to maintain the use of caching of the tiling of vision below a storage limit and to keep the free space of storage above a limit of space free. The garbage collector constantly scans the cache to accumulate size and age statistics for the tiling of vision. When the size of the cache needs to be reduced, the garbage collector selects the newly accessed vision tiling to be removed until the cache size is within limits. The garbage collector runs at a low priority to minimize interference with more critical system tasks. The free storage space limit is designed as a self-protecting limit to prevent the system from running out of storage. The free space limit is verified on a periodic basis, and if it exceeds, the garbage collector becomes a critical system task and operates at a high priority until the free storage space is greater than the free space limit. The files of digital documents can be stored in the Web server or in another server within the network. The image files are typically handled as digital documents with attribute information that is stored with the documents in the document repository 90. The digital document handling system can run independently or together with the image viewing server software. The digital document file can be an image file by frame or a document file without a frame. If the digital document file is in a non-screen format, then it is made as a monochromatic screen image, in gray scale, or in color for the view. The graphical Web browser at the client workstation 10 receives HTML data from the image viewing server 210, which contains hyperlinks for the tiling of vision within the cache 60 of the tiling of vision to be displayed and formatted, which describes the representation of the tiles to form the image vision. The Web browser should initially bring each 220 tiling of vision for a view of the vision server. After the initial view, any view is covered with a preview at the same scale, the Web browser preferably retrieves vision tiles that have been previously displayed from the local cache 210 of the Web browser instead of the server. The operation and utility of viewing documents can be increased by using progressive representation of tiled images. When using an image file format that allows a rigid view of the image to be displayed, while the rest of the content of the image is downloaded, a rigid view of the document can be seen more quickly. Since most web browsers can only transfer 1 to 4 GIF images at a time, usually not all the tiling of vision in the viewing arrangement can be progressively displayed at the same time. Therefore, it is preferred that to implement a progressive deployment, algorithms are provided in the client to accept an alternative data format that can allow the entire document to be seen in the area of the screen to take advantage of progressive deployment, while continuing Taking advantage of the benefits of tiling and caching in the client. This can be achieved in a Web browser environment using algorithms written in Java, JavaScript, or ActiveX technologies. By using client software to improve customer vision, additional improvements can be made to the operation using alternative formats of tiling image and image compression algorithms. An important example can be to use a Portable Network Graphics (PNE) format with the optimization of having the image viewing server and the client transfer only in a common head that will be shared by all the vision tiling and then send the Low resolution compressed image data for each tiling of vision followed by full resolution image data for each viewing tiling. In addition to deploying the vision tiling arrangement for the requested tiling, the HTML data may include hyperlinks used to form a toolbar that provides the user with a way to send requests to the server to change the current view by increasing or reducing the scale of view, shifting the viewing area, or changing the size of the vision. These hyperlinks can refer to the current track and also contain operating codes that direct the image viewing server to form a new view based on the hyperlink selected by the user. The method, apparatus and software of the invention provide an improved client-server architecture using a graphical Web browser to access the image viewing server, which makes the use of the network efficient. By using a tiling and image caching architecture according to the preferred method, relatively small amounts of data need to be transmitted when the user selects a new view of an already received and formed image. The server sends the requested image in the requested format to the workstation and then allows the image to be viewed from the local copy of the image file. The image viewing server provides a better solution using lower cost graphic Web browsers for the workstations to access a network image viewing server that provides views of the image that can be displayed on the workstation by the Graphic Web browser. For example, a raster image file of an engineering drawing of size E is 8 million bytes in size when it is formed into a monochromatic image at 200 pixels by 2.54 cm. With the compression of commonly used data, the image file can be reduced to 250 kilobits. With a modem network connection of 28.8 kilobauds of low bandwidth, with approximately 3 kilobits per second of output, it takes 83 seconds (250 KB / 3KB / second) to transfer the image file to the workstation application in order to be able to see it With the image viewing server only the image data that will be displayed need to be transmitted. A typical view size of 896 by 512 pixels is developed in a 7 by 4 array of 128-pixel, 128-pixel viewing tiling. The monochromatic tiling of vision is transmitted in a compressed format that typically produces tiling that is 512 bits each, so that the whole view is approximately 14 kilobits (0.5 kb x 28 tiling) and the transfer takes approximately 4.8 seconds (14 KB / 3 KB second). This method of image viewing provides a better response to the user with a much lower demand on the network connection. A local area network typically uses a medium of 10 megabits per second, so that the savings of the efficiency of the image viewing server are not obvious. However, if the network of 10 megabits per second is shared by 100 users then the average bandwidth per user is only about 12.5 kilobits per second, so that the efficiency of the image viewing server is still a benefit. Another benefit of the image viewing server is that the data transfer size remains constant even if the size of the viewing image is increased. If the size of the image file was 4 times larger than the previous example, as may be the case with a larger image, a higher resolution image or a less compressible image, then the network load by the image viewing server it could remain unchanged, while the network load of the traditional image viewer could be quadruple. The invention reduces the software purchase costs, since a browse Web graph is a common workstation application that is not expensive and the expense for web browsers is usually a cost that is required for the workstation to other purposes and in many cases it is included with the operating system of the workstation. Costs for the server software can be shared over hundreds of client workstations- The administrative cost of the software is reduced by using the graphical Web browser which is a common workstation application that is already being administered, eliminating an application from additional proprietary workstation software, and using server software that only needs to be managed on a central server instead of each workstation. The Graphic Web Browser is available on all common workstation types as well as on other devices such as notebook-type computers, pocket computers, network computers and Web TV adapters to provide a widely available solution. It will be apparent to those skilled in the art that various modifications and variations may be made to the network image viewing server and the image viewing server methods without departing from the scope or spirit of the invention. Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention described herein. It is intended that the specification and examples be considered only as illustrative, with a true scope and spirit of the invention being indicated by the following claims.

Claims (27)

1. - A method for identifying and supplying a graphic image from a computer network file server, which comprises providing a network file server where digital document image files are stored, the server adapted to receive requests from a computer. Web browser in a uniform resource locator code (URL), to identify the image file and format selections that are requested, to form the requested view in a grid of viewing tiling, and to transmit the HTML code for the tiling of vision to the Web browser that you request.

2. The method according to claim 1, wherein the URL view requests to identify the image file, the page number within the image file, formats comprising one or more of the position, scale, rotation angle, mirror x, mirror and, investment, region, updated to select an updated view, and a vision template to determine the presentation of the view, and where the transmitted HTML code for the tiling of vision causes the Web browser that requests deployment the requested view of the graphic image requested in the requested format.

3. - The method according to claim 2, wherein an arrangement of vision tiling selected from the viewing tiling grid satisfies the request described by URL and is calculated by the network file server and transmitted over the network to the network. Web browser on the workstation in an HTML and GIF format.

4. The method according to claim 3, wherein the file server creates an HTML output file that lists the size, position and hyperlink of each viewing tiling that will be displayed, and where the tiling of vision is stored. in a GIF image file format for each vision tiling to be displayed by the Web browser.

5. - The method according to claim 1, wherein the adjacent tiling of the view tiling arrangement transmitted to the Web browser fulfills the request described by URL and is calculated or computed by the server during a period subsequent to the transmission of the provision of tiling of vision, and the adjacent tiling of vision are stored in cache in the server in anticipation of possible future requests that will be received from the workstation.

6. The method according to claim 5, wherein the tiling of vision adjacent to the provision of tiling of vision transmitted to the Web browser are computed by the server in a spiral sequence clockwise or vice versa clockwise outside the disposition of transmitted tiling of vision and the adjacent vision tiling are transmitted in cache in the network server memory.

7. The method according to claim 2, wherein the web browser is adapted to cache the tiling of vision in the memory of work station, which have been received to display and view, and in the case of which a subsequently requested view including one or more of the caching tiling in the workstation memory, to retrieve one or more caching tiling from the local workstation memory instead of the server net.

8. The method according to claim 2, wherein the tiling of vision are GIF image files of 128 pixels by 129 pixels.

9. The method according to claim 2, in each vision tiling is identified by the server with a unique URL that includes storage location and appointment data of vision tiling.

10. The method according to claim 2, wherein a separate storage subdirectory is computed by the server in the vision tiling cache for each image.

11. The method according to claim 9, wherein the URL includes identification data for each vision scale through the file name for each vision tiling.

12. The method according to claim 11, wherein the tiling of vision are named in the V <format; SCALE > < TILE_NUMBER > .GIF, where the value < SCALE > is a two character strip formed from the base coding 36 of the vision scale number as expressed in parts by 256; the value < TILE NUMBER > is a strip of 5 characters formed from the base coding 36 of the number of tiling as determined by the formula: TILE_NUMBER = TILE_ROW * IMAGE_TILE_WIDTH + TILE_COLUMN; wherein the values TILE_ROW and TILE_COLUMN begin at 0.

13. The method according to claim 11, wherein the angle of rotation of the view, the mirror image x of the view, the mirror image and the view and the inverted view information is encoded in the URL naming format in the form of V < SCALE > < TILE_NUMBER > < VIEW_ANGLE > .X_MIRROW > < Y_MIRRO W > < INVERT > .GIF, where the value VIEW_ANGLE is in the form of A < ANGLE >; and X_MIRROW, Y_MIRROW, and INVERT are encoded by individual characters X, Y, and I respectively.

14. The method according to claim 1, wherein the Web server is implemented with a Web Server Common Gateway Interface (CGI).

15. The method according to claim 1, wherein the request client workstation is configured to send small or handwritten applications to improve communications with the network image server to further reduce the amount of redundant data that is required. They need to be transferred to the client's workstation.

16. The method according to claim 1, wherein the client workstation and the network server are adapted to minimize the amount of redundant data that is sent to the client.

17. The method according to claim 1, wherein higher resolution images are used by the server to sample lower resolution images in the client.

18. The method according to claim 1, wherein the server is adapted to provide view scaled images in gray and / or color to the client.

19. The method according to claim 1, wherein the files without a frame are provided with a frame through the server to be seen by the client.

20. The method according to claim 1, wherein the image files are stored in a second server and recovered and temporarily stored by the network server when necessary.

21. An apparatus comprising a computer network server adapted to store image files of digital documents, programmed to receive requests from a client Web browser in a URL code, the URL specifying a view, which identifies an image file and a format, to form the requested view, and to transmit the HTML code for the client Web browser to display this.

22. The apparatus according to claim 21, wherein the server is programmed with Web server software and a former viewer, which is adapted to interpret the vision request, calculate or compute a grid of tiling vision of the image, compute a disposition of tiling of vision that is necessary for the required view, form the disposition of vision tiling necessary for the required view, and create an HTML output file to describe the composition of the view to the Web browser.

23. The apparatus according to claim 21, wherein the server is programmed with a background vision former adapted to compute and cache, during periods between view requests, tiling of vision surrounding the vision request. most recent on the same scale as the most recent view, so that when a new displaced view is subsequently requested, the server program computes which of the surrounding cached surrounding view tiling can be used in response to a request for a view displaced, and forms a view comprising one or more caching tiling.

24. The apparatus according to claim 21, wherein the surrounding vision tiling is computed by the background view former in a spiral clockwise or counterclockwise out of sight more recent.

25. The apparatus according to claim 21, wherein the client Web browser is programmed to cache the tiling of vision, which is received in response to requests to the server and to access the tiling of vision in cache when they are included in a provision included in a subsequently requested hearing. 26.- The apparatus according to claim 21, wherein the server is programmed with a garbage collector in vision tiling cache adapted to remove the newly accessed vision tiles when the maximum storage distribution limits or limits are reached. minimum free storage space limits. 27.- An article comprising a computer program recorded in a magnetic or optical medium for use in a network server, comprising a code that interprets HTTP requests from a workstation for a particular view of a document image file stored in memory, retrieves the image file of digital documents, forms a grid of tiling of vision corresponding to the requested vision of the image, and calculates the HTML code for the grid of tiling of vision in a form that can be transmitted from the server to the workstation.