IIPMooViewer is a high performance light-weight HTML5 Ajax-based javascript image streaming and zooming client designed for the IIPImage high resolution imaging system. It is compatible with Firefox, Chrome, Internet Explorer (Versions 6-10), Safari and Opera as well as mobile touch-based browsers for iOS and Android. Although designed for use with the IIP protocol and IIPImage, it has multi-protocol support and is additionally compatible with the Zoomify, Deepzoom, Djatoka (OpenURL) and IIIF protocols.
Version 2.0 of IIPMooViewer is HTML5/CSS3 based and uses the Mootools javascript framework (version 1.5+).
- Fast and light-weight
- Pan and zoom of ultra high resolution imaging
- Multi-protocol support: IIP, Zoomify, Deepzoom, Djatoka (OpenURL) and IIIF protocols
- Image rotation
- Mobile device support: iOS and Android
- HTML5 Fullscreen API support
- Annotations of regions of images
- Synchronized viewer capability
- Image blending comparison
- Localization
The distribution contains all the necessary library files in both compressed and uncompressed formats. Modify the parameters in the index.html template example file provided.
You first must have a working version of the IIPImage server running if you want to use the IIP protocol and features. See http://iipimage.sourceforge.net for details. IIPMooViewer, however, also supports the Zoomify and Deepzoom protocols if you are unable to install the server or are working in a legacy environment.
Create a pyramidal tiled TIFF image using VIPS (http://vips.sf.net) or imagemagick. Or JPEG2000 if you have a JPEG2000 enabled IIPImage server.
Now modify the path and image names in the example HTML page provided - index.html - to create a working client :-)
var iipmooviewer = new IIPMooViewer( "viewer", {
image: "/path/to/image.tif",
credit: "My Title"
});
Note that IIPMooViewer works best when it's in its own page, as it expands to
100% width and height. If you wish to embed it in another page, you may do so
with an <iframe>. To get IIPMooViewer's fullscreen to work when it is in an
<iframe>, add the allowFullScreen attribute to the <iframe> tag.
/javascript : the necessary minified iipmooviewer and mootools javascript files
/css : iip.css and ie.css (for Internet Explorer)
/images : icons and image files
/src : uncompressed source javascript files
Minified files are created with the closure compiler with the following command:
java -jar /path/to/compiler.jar --js src/mootools-more-1.5.1.js src/iipmooviewer-2.0.js src/navigation.js src/scale.js src/touch.js src/protocols/iip.js src/annotations.js src/blending.js src/lang/help.en.js --js_output_file js/iipmooviewer-2.0-compressed.js --compilation_level SIMPLE_OPTIMIZATIONS
You can thereby customize your build to include only those components you need. For example, if you require Zoomify or do not require annotation support, simply add or remove these from the build. Make sure, however, that the core iipmooviewer js file is included before the other components. Otherwise, if you prefer not to rebuild, use the default build and add the extra components you want as extra includes.
Options to the IIPMooViewer class constructor: (The only obligatory option is the image variable)
image : The full system path to the image. On Windows as on other systems this should be a UNIX style path such as "/path/to/image.tif". Note that this is an absolute system path and not the path relative to the webserver root
server : The address of the IIPImage server. [default: "/fcgi-bin/iipsrv.fcgi"]
credit : a credit, copyright or information to be shown on the image itself
render : the way in which tiles are rendered. Either `random' where the tiles are fetched and rendered randomly or 'spiral' where the tiles are rendered from the center outwards [default: "spiral"]
showNavWindow : whether to show the navigation window. [default: true]
showNavButtons : whether to show the navigation buttons on start up: true or false [default: true]
navWinSize : ratio of navigation window size to the main window. Wide panoramas are scaled to twice this size [default: 0.2]
navigation : a hash containing options for the navigation box: (a) id: the id of the element where the navigation box will be embedded. Defaults to the main container. (b) draggable: a boolean, indicating whether the navigation box is draggable. Defaults to true, however, if a navigation id is specified, defaults to false. (c) buttons: an array of the available buttons: reset, zoomIn, zoomOut, rotateLeft, rotateRight. Defaults to: ['reset','zoomIn','zoomOut']
showCoords : whether to show live screen coordinates [default: false]
scale : adds a scale to the image. Specify the number of pixels per unis. Should be given in pixels per mm if the default meter unit is used. Otherwise pixels per degree if degree units are used.
units : define the units used. Can be in degrees or meters [default: meters]. Or define completely new unit system. The default structure is the following:
{
dims: ["pm", "nm", "µm", "mm", "cm", "m", "km"], // Unit suffixes
orders: [ 1e-12, 1e-9, 1e-6, 0.001, 0.01, 1, 1000 ], // Unit orders
mults: [1,2,5,10,50,100], // Different scalings usable for each unit
factor: 1000 // Default multiplication factor
}
prefix: path prefix if image subdirectory moved (for example to a different host) [default: "images/"]
enableFullscreen : allow full screen mode. If "native" will attempt to use Javascript Fullscreen API. Otherwise it will fill the viewport. "page" allows fullscreen but only in viewport fill mode. False disables. [default: "native"]
winResize : whether view is reflowed on window resize. [default: true]
viewport : object containing center x, y, resolution, rotation and contrast of initial view. For example, to start at resolution 4 with the center of the view port at both 90% of the size of the image:
{resolution:4, x:0.9, y:0.9, rotation:0}
protocol : protocol to use with the server: iip, zoomify or deepzoom [default: "iip"]
preload : preload an extra layer of tiles surrounding the viewport [default: false]
annotations : An object containing object structures with parameters "x", "y", "w", "h", "title", "text", "category" where x, y, w and h are the position and size of the annotation in relative [0-1] values, title is an optional title for the annotation, category is an optional category for the annotation and text is the body of the annotation
getRegionURL(): If using the default IIP protocol, this functions returns the IIPImage server URL needed to export the region of the image within the view port as a single image. Thus, to export the current view, call this function and use the result as the source of an image. This example exports, when the user presses the "p" key, the view into a new window which can then be saved as a whole image.
window.addEvent('keypress', function(e){
if( e.key == "p" ) window.open(iipmooviewer.getRegionURL());
});
rotate(x): Rotate the view by x degrees
moveTo(x,y): Move the view to position x,y at the current resolution, where x,y are the coordinates of the top left of the view port.
centerTo(x,y): Center the view at relative position x,y at the current resolution, where x,y are resolution independent coordinate ratios (0.0 -> 1.0 ) of the center of the view port.
zoomIn(): Zoom in by a factor of 2
zoomOut(): Zoom out by a factor of 2
zoomTo(z): Zoom to a particular resolution, z
setCredit(t): (Re)set the text in credits to the HTML given by t
recenter(): Center our view
reload(): Reinitialize our view, using the initial viewport settings if given during class initialization
changeImage(i): Load a new image, i, and reinitialize the viewer
toggleNavigationWindow(): toggle the navigation window
toggleFullScreen(): toggle fullscreen mode
toggleAnnotations(): toggle display of annotations
You can supply a list of annotations for the image which will be overlaid while
navigating the image. These must be supplied in an object containing a list of
individual annotation objects, each with parameters describing the size,
position, title, category and text. The size is set by the w,h properties, and
the position is described by the x,y properties (top-left offset). The size and
position (obligatory) parameters are ratios (i.e go from 0.0 to 1.0) of the
image's dimensions. The text parameter provides the content of the annotation
and can contain any valid HTML, which can be styled normally via CSS. All
annotations are created as divs of class annotation.
For example:
var annotations = {
1: { x: 0.7, y: 0.6, w: 0.2, h: 0.18, category: "pigments", text: "prussian blue" },
2: { x: 0.1, y: 0.8, w: 0.15, h: 0.1, category: "pigments", text: "azurite" },
3: { x: 0.7, y: 0.4, w: 0.1, h: 0.1, category: "people", text: "Mary" }
};
The 1,2,3 are unique IDs which can be either numbers or strings.
The created element's ID is the annotation's ID with annotation- prepended.
Categories are ways of creating groups of annotations and the category will be added to the class. Thus for a category of, for example, 'retouches' the annotation divs will be of class 'annotation retouches', allowing you to access these via a class selector. So, for example, to set the colors of these differently to the others, simply with javascript, use a selector:
$$('.annotation.retouches').setStyle('borderColor', "blue")
or in CSS:
.annotation.retouches{
border-color: blue;
}
Annotation editing is possible by including the annotations-edit.js file, which extends the IIPMooViewer class. The function newAnnotation() creates a new blank annotation in the centre of the view. Double click on any existing annotation to move, resize of modify it. When an update occurs, an annotationChange event occurs, which can be captured and used to send the results back to a server via an AJAX call.
For example, to send the updated list of annotations back to annotations.php:
iipmooviewer.addEvent('annotationChange', function(action, annotation_id) {
var data = {id: annotation_id, action: action};
// `action` is either `updated` or `deleted`
if (action == 'updated') {
// If the annotation has been updated, send the updated data.
data.annotation = JSON.encode(this.annotations[annotation_id]);
}
var metadata = new Request.JSON({
method: 'post',
url: 'annotations.php',
data: data
}).send();
});
It is possible to synchronize two or more instances of iipmooviewer, so that they will zoom, pan and rotate at the same time. To do this, simply create your viewers and synchronize them together using the IIPMooViewer.synchronize() function, which takes an array of viewer instances. For example:
// Create viewers
var viewer1 = new IIPMooViewer( "viewer1", {
image: 'image1.tif'
});
var viewer2 = new IIPMooViewer( "viewer2", {
image: 'image2.tif',
showNavWindow: false, // Only show navigation window on first viewer
showNavButtons: false
});
// Synchronize our viewers
IIPMooViewer.synchronize([viewer2,viewer1]);
IIPMooViewer fires the 'load' event when it has fully finished loading. To attach to this event, use the addEvent function:
iipmooviewer.addEvent('load', function(){
// do something
});
IIPMooViewer supports the IIP, Zoomify, Deepzoom, Djatoka and IIIF protocols. By default it will use IIP, but to use, for example Zoomify, first include the protocol after the other javascript includes as this is not included by default:
<script type="text/javascript" src="src/protocols/zoomify.js"></script>
Protocols such as zoomify don't have a server as such, so set this value to / or to the path prefix for the image. For example, if you have an image with URL /images/zoomify/image1, use:
new IIPMooViewer( "viewer", {
server: "/images/zoomify/",
image: "image1",
protocol: "zoomify"
});
Note that for Deepzoom, the image name should be name of the .dzi or .xml Deepzoom index file.
A number of localizations exist in the src/lang folder. To use, simply include the desired language file as an extra javascript include. To create a new localization, create a new or modify an existing localization file and include this extra javascript file.
For example to create a new Chinese localization, create the file lang/help.zh.js and include it after the main iipmooviewer includes:
<script type="text/javascript" src="javascript/mootools-core-1.5.1-full-nocompat-yc.js"></script> <script type="text/javascript" src="javascript/iipmooviewer-2.0-compressed.js"></script> <script type="text/javascript" src="src/lang/help.zh.js"></script>
It's also possible to load several images for comparison and dynamically blend between them to compare. This is useful, for example, for comparing scientific imagery of the same scene or object. Images should be of the same size and registered. The blending component is in src/blending.js, but is built by default into the main compressed iipmooviewer js file in the javascript/ folder. To use, simply load the the viewer as normal with the default image, but use the blend() function to provide a list of all images and text for use in the selection box. For eample:
var iipmooviewer = new IIPMooViewer( "targetframe", {
image: 'color.tif',
credit: 'Compare scientific images'
});
iipmooviewer.blend( [ ['color.tif','color'],
['uv.tif','ultra-violet'],
['ir.tif','infra-red'],
['xray.tif','X-ray']
] );
A hash tag in the form x, y, resolution can be appended to the URL to link to a particular area within the image at a particular resolution. x and y should be resolution-independent ratios from 0.0 -> 1.0 and the resolution an integer representing the desired resolution number (where 0 is the smallest resolution). For example:
http://your.server/iipmooviewer/test.html#0.5,0.5,5
will set the initial view of the image to the x,y coordinate 0.5, 0.5 (the center of the image) at resolution number 5.
The hashchange event is also used if supported by the browser to update the view if the coordinates change. This can be used, for example, to maintain a sequence or history of view changes.
Use control-c in order to obtain the tagged URL of the current view.
It is possible to restyle widgets such as the credit information box. Simply create your own CSS rule to override the default rule for:
.iipmooviewer .credit
to change the font, background or borders etc.
Please refer to the project site http://iipimage.sourceforge.net for further details
(c) 2007-2014 Ruven Pillay