Home - Software Reviews PagePlus Versions Review WebPlus Versions Review DrawPlus Review Freeware 7-Zip Archive Program BabelMap Unicode Character Map Chattha Sangayana Tipitaka Exam Diff Text File Comparison Tool FastStone Viewer and PhotoResizer IrfanView Image Viewer and Converter Jarte - A Wordpad Clone High-Logic MainType  - Font Manager LibreOffice Office Suite Free OpenType Fonts Microsoft image Composie Editor PDF-XChange Editor Review Vivaldi Browser Review Shareware FastStone Capture Screen Grab Utility High-Logic FontCreator - TrueType/OpenType Font Editor High-Logic Scanahand - Create Fonts Automatically from Scans

Microsoft Image Composite Editor

DownloadSupportRelease Notes

Released 24th February 2015

I have tried PanoramaPlus X4 and Scan, Stitch, and Share, but neither comes close to the speed, power, or easy of use of this freeware. Microsoft’s Image Composite Editor may be development software, but it seems to be rock-solid. If you have a 64-bit Operating System, the ability of the 64-bit version to access all of your PC memory makes this the obvious choice. I am sure that the 32-bit version will be fine for less demanding tasks.

I don’t have any video to try stitching, so I will give a brief run-through of the simple process of stitching two or more images. The second example stitches together ten screen shots from the Site Structure view of this Softer Views web site to make a single large image. The first illustrates the process of stitching two camera images. There are lots of options for tweaking the projection, but I will just demonstrate with the simplest Cylindrical Projection.

New Panorama From Images

Import

To get started, click on “New Panorama from Images” on the opening screen, and select two or more camera images that you wish to stitch together. This will import the images and display them in the Import window.

If there are many images they can be sorted, but in this case, since there are only two images, there is nothing more to be done.

Assuming you imported the right images, just click on Stitch.

Stitch

The imported images will be stitched automatically using the default Cylindrical Projection. The two test images are 5184 x 3456 pixels (almost 18 Megapixels), so fairly typical of modern camera images. The stitching process took just 2 or 3 seconds on my 2½ year old hardware:

After stitching, the projection can be changed, or adjusted using the arrow cursor on the grid, and the zoom slider can be used to zoom in. I adjusted the projection so that the sides of buildings in the focal point were vertical.

Crop

When you’re satisfied with the projection, click on crop to go on to the next stage. Zoom out to view the entire picture, and drag the side handles to resize the crop rectangle. At this stage, I recommend leaving as much of the original picture as you can, as long as it’s only going to involve some simple cloning of the sky to fill in the missing areas. It is easy enough to crop again later in your image editor.

In this picture, I wanted to keep all three windows of the church, and at least part of the dome on the adjoining tower. The sky area is going to be very easy to fill in with the clone tool.

The crop rectangle can be freely resized with the corner handles, and the image can be panned around the screen with the Hand cursor to bring each corner of the image into view while you make fine adjustments to the crop area.

When I did this, I found that the bottom right corner was going to require too much work with the clone tool, so I accepted a compromise, losing some of the third window on the church to reduce the work needed for cloning the parked car.

If something doesn’t look quite right with the projection, just click on the back button to refine the projection until you’re satisfied, then click on crop to resume the crop operation.

After adjusting the projection a little, you might need to crop less of the stitched image.

Another powerful option is to use the Autocomplete Option on the Crop window. This basically does the cloning for you, filling in those gaps at the edge of the picture with pixels cloned from the adjacent areas. This works very well for areas of sky or ground, but not so well for other details. Mouse over the image above to see the effect of Autocomplete — note the area to the right of the church tower.

Export

The final stage is to export the stitched image to an image format of your choice. If it’s a photograph, then JPG is obviously the best compromise between file size and image quality. The default setting of High Quality (75) may be too low for photographers who want their work to look its best. Open the drop down to select the Superb Quality (90). Or use PNG doing the compression to JPG later.

If the images are not photographs, but screen shots, as in my second example, then choose PNG as the File Format, and check that the scale is set to 100%. There is no compression setting for PNG images — the default compression is not bad, but it can be improved by opening the resulting image in FastStone Capture and resaving it at maximum compression.

Creating a Panorama from Screen Shots

If you want to stitch several images from a mapping program, all you need is to make sure that there is sufficient overlap between the screen shots. Save images in PNG format, and Export to PNG after stitching them to retain maximum quality

For this demonstration, I took ten screen shots of the site structure view of this SofterViews web site in WebPlus X8. I then stitched the shots together to make a single large panorama of the entire site structure.

After importing the images, I had to manually choose the Planar Motion option for the camera, or the stitch would not work at all using the auto-detection mode.

After clicking “Stitch,” the pictures were then stitched in the correct way. Then I could go on to crop the image and export it to a PNG image. I reduced the number of colours to 256 using IrfanView to reduce the file size of this 5991 x 1160 pixel image to just 429 Kbytes. I used the PNGOUT plugin in IrfanView to reduce it to 391 Kbytes.


Page last updated on 29 June 2017