Patent Grant
7792862
A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by any one of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever
A computer program listing appendix is submitted on a compact disc containing files named Tcl-based applet, created 17 Apr. 2006, of size 6.0 KB and named Exemplary object index, image map and secondary image map, created 17 Apr. 2006, of size 46.1 KB. and the compact disc is hereby incorporated by reference.
The present invention relates to a method and apparatus for identifying features of multidimensional image data in hypermedia systems.
The World Wide Web (“Web” or “Internet”) provides a simple mechanism, called an image map or ISMAP, for linking two-dimensional spatial data (e.g., images) to related symbolic information such as Universal Resource Locators (URLs). Image maps are a simple technology that link simple polygonal regions within images, often referred to as hotspots (e.g., a graphically defined area in an image), to the locations of data objects on the Internet via a hyperlink. Web image maps currently are the standard mechanism used for creating graphically attractive user interfaces to Web pages. For example, an image could be made into an image map by assigning hotspots to each region of interest on the image. Unfortunately, the standard polygon-based ISMAP technology used in most Web image map systems can only work with simple polygon maps, and becomes either intolerably slow or totally unusable for mapping high-resolution images with large numbers of irregularly shaped objects, such as, for example, objects in the medical anatomy image data which comprise the National Library of Medicine's Visible Human Project.
U.S. Pat. No. 4,847,604, which is hereby expressly incorporated by reference, describes a method and apparatus to provide additional information concerning a feature of a displayed image by pointing to the location of the feature. The U.S. Pat. No. '604 provides for the correlation of particular locations in the image with the additional information for two-dimensional images. A need exists to provide such correlation for multidimensional images.
A multi-dimensional object indexing system allows many discrete objects to be mapped within a single multi-dimensional dataset. A secondary spatial image of an original image is correlated using a multi-dimensional coordinate value to provide an object index for each object defined in the original multi-dimensional image.
At 1030, a multidimensional coordinate location is determined for the location identified at 1020. For example, the image map may be a two-dimensional representation (e.g., x-y) of a three dimensional image, such as a three dimensional medical anatomy image maintained by the National Library of Medicine's Visible Human Project. The two-dimensional representation may reflect a particular third dimension view. For example, a particular slice of the image may be shown to a user thus providing a predetermined third dimension, or z-value, for the representation of the image displayed to the user. Accordingly, selecting a particular location on the representation defines the x and y values for the location and the third dimension is already defined. As a result, the multi-dimensional coordinates of the selected image location can be determined.
In an alternative exemplary embodiment of the present invention, the selection of a image for correlation with additional information may be a multi-step process. For example, an image of an entire anatomical body could be presented to a user and a general area of the image selected, such as the chest, head or abdomen. In response, a detailed view of the related region could be displayed to the user. The user would then select a particular location in the image and receive additional information related to the selected location. As noted above, the images presented to the user will generally be two-dimensional representations viewed on a predetermined or selectable third dimension. In addition, the original image can be manipulated, rotated or sliced for viewing in a preferred orientation. The image location eventually selected by a user, regardless of the orientation and display of the image, will correspond to a particular multi-dimensional coordinate which can be correlated with additional information as described below.
Alternatively, the image map could be a video clip having x and y values for each frame of the video clip and the third dimension, in this case time, would be determined by the particular frame of the video clip. Thus, multi-dimensional values can be determined for an indicated location. The present invention is not limited, however, to three dimensions as four or more dimensions can be processed according to exemplary embodiments of the present invention. For example, a three dimensional image dataset that changes through time would provide for four dimensions (e.g., x, y, z and time).
At 1040 a secondary map is queried for the determined multi-dimensional value where, for example, a given x,y,z coordinate in the original image data corresponds to an homologous x,y,z location in the secondary “map” of, for example, 24-bit voxels (e.g., three-dimensional pixels). The 24-bit voxel found at that location in the secondary map then acts as a unique 24-bit object index for the image feature in the original data. At 1050, the system then performs a database table lookup for that 24-bit index in order to find the related descriptive text information or “hot data,” which may be in the form of, for example, HTML code, a URL pointer to a remote Web resource, or Tcl-based applet code. Using the 24-bit voxel allows as many as 16 million discrete objects to be mapped within a single three-dimensional dataset. If desired, even more objects could be mapped in the dataset by using 32-bit voxels. Using an exemplary embodiment of the present invention, animated GIFs, AVIs and MPEGs can be processed to provide enhanced functionality to a Web page to extend image capabilities to multi-dimensional datasets.
At 2040, a secondary image map is generated for the image data. The secondary image map provides, for example, a spatial index of the image with equivalent spatial dimensions as the original image. The secondary image map utilizes, for example, unique 24-bit voxels to associate a unique 24-bit identifier with the pixels in an object outlined at 2020. Alternatively, pixels within an object can have different unique identifiers defining, for example, sub-regions in the object having associated program actions. Other object indices could be used instead of the 24-bit voxels, however, such as a list of polygons or some other indice that correlates a coordinate value in the original image with the secondary image map. The 24-bit values are, for example, arbitrarily assigned to groups of pixels or individual pixels in the object.
According to an exemplary embodiment of the present invention, each pixel in the object can be an independently addressable hotspot, including for multi-dimensional images. For example, for each mapped multi-dimensional image on a Web page, the Web server hosting the Web page stores the secondary image map. Each pixel or group of pixels in a hotspot defined for the image corresponds to, for example, a flat area of 24-bit color in the secondary image map, as each object pixel or group of pixels owns a unique 24 bit color in the secondary image map. While the primary (original) image is displayed to the user, the secondary image may is generally not shown to the user, although there may be circumstances in which it is desirable to show the secondary image map. At 2050, object table files are generated to associate the pixels in each object with the program action identified at 2030. The object table can have the form of, for example, a database lookup table indexed by the 24-bit voxels to find the corresponding program action for each pixel.
Exemplary operation of the method for identifying features in multi-dimensional image data is as follows and as illustrated in
As described above, the third dimension may be predetermined as a function of the image presented to the user such as a particular “slice” of an image or a frame of a video clip. Once the multi-dimensional coordinate value is determined, the Web server 3020 launches a, for example, Java-based CGI program to find the corresponding multi-dimensional coordinate value in the secondary image map to determine the unique 24 bit value corresponding to the pixel clicked on in the image by the user. The CGI program then sends the 24-bit value to a, for example, Tcl-based server which takes the 24-bit value and fetches a record from the database lookup table for the corresponding program action to be taken, or the Web server 3020 itself processes the 24-bit value and uses the value as an index into database 3040 coupled to Web server 3020 to determine the program action to be performed. The program action is relayed back by the Web server 3020 for display to the user at client computer 3010, usually in a frame next to the image map. The program action can include, for example, browser redirection, encapsulated HTML, dynamic HTML and downloading and running full-frame applets in the user's Web page. Other program actions can be performed as well.
Thus, according to an embodiment of the present invention, individual pixels or voxels in objects of multi-dimensional image data can be automatically linked to desired program actions in contrast to prior art systems which are limited to indexing of two-dimensional Web images.
The additional dimension added to the indexing of image maps can vary as a function of the type of image data displayed. Thus, for example, the additional dimension could be the z-coordinate for three-dimensional images or the time dimension for video clips or both for three-dimensional image data recorded over time as well as additional dimensions. Like the multi-dimensional image data, the secondary image map according to an exemplary embodiment of the present invention can have n-dimensions. In addition, the individual pixels in an object of a multi-dimensional image can be individually addressable. Moreover, by adding a dimension to a standard image map request protocol in accordance with an exemplary embodiment of the present invention, the display of an image on a client side of computer network can be independent of the server performing the object identification in the image, thus allowing the capabilities provided by the present invention to be implemented using conventional personal computers or even less powerful devices such as handheld computers or smartphones.
An exemplary implementation of the method according to an embodiment of the present invention is illustrated below and uses a small (30K) Tcl-based client applet which downloads and runs within the user's Web browser page. For example, when a user accesses a web page with multi-dimensional image data to be viewed, an applet such as the applet described below would be automatically downloaded to the client computer. The applet would then, for example, fetch the appropriate image data, secondary image map and object index table and download the data to the client computer. The client computer web browser then interacts with the download applet to view the image data, for example in the manner described above. For example, the Tcl applet converts the image data back to a viewable image. If the image is a video clip, for example, then the video clip would be displayed to the user. The applet program then waits for the user to click on a location in the video clip. When the user clicks a location, the frame number is determined and the x-y location of the click is determined as described previously, thereby enabling a multi-dimensional coordinate location for the click to be determined to use as an index into the secondary image map, from which the program action can be determined via an object index table.
Computer program listings for an exemplary Tcl-based client applet and exemplary object index, image, and secondary image map are included in appendices on the above-described compact disc.
An alternative exemplary implementation according to an embodiment of the present invention could use a small (30K) Tcl-based client applet similar to that described above which downloads and runs within the user's Web browser page but where the image data is not downloaded to the client computer but instead is remotely manipulated by the user. For example, the image data could be too large to download to the user. Such an applet would, for example, open a socket communications channel to a remote visualization server that stores and manipulates three dimensional image data. When the user makes a change to the controls in the applet GUI, such as to specify rotation around an axis, the remote visualization server would perform that computation on the data and transfers the resultant data display back to the user for viewing the result of the manipulation within the display frame of the client applet. The user could, for example, slice through the data at any angle and click upon any voxel on any slice surface to cause the associated descriptive text to be shown in the lower window of the applet, for example in the manner described above. The efficiency of this type of system is such that object identification and response occurs in nearly “real time” with identification speeds of, for example, approximately ½ second over 56 KB Internet connections.
In another alternate embodiment, given a user specified description, the descriptive information for the various features can be searched to determine whether there is a correspondence with the user specified description, in order to find the features responsive to the user specified description. The physical locations of the features on the displayed image are then indicated. An example method for searching the descriptive information is described in the incorporated-by-reference U.S. Pat. No. '604.
This is a continuation of U.S. patent application Ser. No. 10/613,980, filed Jul. 5, 2003 which is a continuation of U.S. patent application Ser. No. 09/825,570, filed Apr. 3, 2001 (U.S. Pat. No. 6,616,701 issued Sep. 9, 2003), which is a continuation of U.S. patent application Ser. No. 09/316,496, now abandoned, filed May 21, 1999, which claims priority from a Provisional Application Ser. No. 60/086,620, now abandoned, filed May 23, 1998.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4686522 | Hernandez | Aug 1987 | A |
| 4847604 | Doyle | Jul 1989 | A |
| 5096427 | Sadigh-Behzadi | Mar 1992 | A |
| 5708845 | Wistendahl | Jan 1998 | A |
| 5737553 | Bartok | Apr 1998 | A |
| 5940538 | Spiegel | Aug 1999 | A |
| 5982372 | Brush | Nov 1999 | A |
| 5987509 | Portuesi | Nov 1999 | A |
| 5990941 | Jackson | Nov 1999 | A |
| 6057854 | Davis, Jr. et al. | May 2000 | A |
| 6121981 | Trower | Sep 2000 | A |
| 6230167 | Libscomb | Aug 2001 | B1 |
| 20060174209 | Barros | Aug 2006 | A1 |
| 20080232658 | Sugaya et al. | Sep 2008 | A1 |
| Number | Date | Country |
|---|---|---|
| 2298769 | Sep 1996 | GB |
| Number | Date | Country | |
|---|---|---|---|
| 20080163038 A1 | Jul 2008 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10613980 | Jul 2003 | US |
| Child | 12040194 | US |
