This Title All WIREs
How to cite this WIREs title:
WIREs Comp Stat

Image grand tour

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract We describe a method of fusing, visualizing, and exploring multiple registered images called the image grand tour (IGT). The IGT is based on the grand tour idea for exploring high‐dimensional data sets. In this article, we provide information on the general grand tour and show how this can be adapted to explore many linear combinations of the registered images in search of interesting structure and information. We provide several examples where the IGT is used for remote sensing, the search for rock art pictographs, and sensor fusion. WIREs Comput Stat 2013, 5:198–206. doi: 10.1002/wics.1253 This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Exploratory Data Analysis Statistical Learning and Exploratory Methods of the Data Sciences > Image Data Mining Applications of Computational Statistics > Signal and Image Processing and Coding

This shows the six images of the same potential minefield. Note that some mine‐like objects (arranged in a regular pattern) can be seen in some of the bands, but not in all of them. There does not appear to be any mines in the image in the lower right corner. We used the montage function in the MATLAB Image Processing Toolbox to create the montages in this article. Note that none of the component images are cropped, and they represent the entire component used to get the fused image in Figure 2.

[ Normal View | Magnified View ]

This is a projection of two infrared images and one in the visible spectrum. Note that we can see more details in this fused image that cannot be seen in the individual component images. For example, we see some of the wave action along the shore, additional ship details, and several objects highlighted by the arrows.

[ Normal View | Magnified View ]

The image on the left corresponds to the visible spectrum, and the image on the right is one of the infrared images used in the IGT that produced the image in Figure 8.

[ Normal View | Magnified View ]

This shows one of the image projections from the IGT, where the entire figure can now be seen.

[ Normal View | Magnified View ]

This is a conventional photograph of a rock that has a figure painted on it. We can barely see that something is there.

[ Normal View | Magnified View ]

This is one of the fused images from the IGT of the montana Landsat data. Note the appearance of cultivated areas that do not appear in the separate bands.

[ Normal View | Magnified View ]

This shows four of the Landsat TM bands (2, 3, 4, 6) in the montana data set that is included in the Image Processing Toolbox. The multibandread function was used to get the Landsat data into the MATLAB workspace.

[ Normal View | Magnified View ]

One projection from the IGT, where we can see potential mines (shown by the arrows) that are not clear in all of the spectral bands. Other mine‐like objects arranged in a regular pattern (horizontal line) are now visible.

[ Normal View | Magnified View ]

Related Articles

Imaging: An Interdisciplinary View

Browse by Topic

Applications of Computational Statistics > Signal and Image Processing and Coding
Statistical Learning and Exploratory Methods of the Data Sciences > Image Data Mining
Statistical Learning and Exploratory Methods of the Data Sciences > Exploratory Data Analysis

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts