A. Ardeshir Goshtasby

Guest editorial: Image fusion: Advances in the state of the art

Image fusion is the process of combining information from two or more images of a scene into a single composite image that is more informative and is more suitable for visual perception or computer processing. The objective in image fusion is to reduce uncertainty and minimize redundancy in the output while maximizing relevant information particular to an application or task. Given the same set of input images, different fused images may be created depending on the specific application and what is considered relevant information. There are several benefits in using image fusion: wider spatial and temporal coverage, decreased uncertainty, improved reliability, and increased robustness of system performance. Often a single sensor cannot produce a complete representation of a scene. Visible images provide spectral and spatial details, and if a target has the same color and spatial characteristics as its background, it cannot be distinguished from the background. If visible images are fused with thermal images, a target that is warmer or colder than its background can be easily identified, even when its color and spatial details are similar to those of its background. Fused images can provide information that sometimes cannot be observed in the individual input images. Successful image fusion significantly reduces the amount of data to be viewed or processed without significantly reducing the amount of relevant information.

On the Canny edge detector

Abstract The Canny edge detector is widely used in computer vision to locate sharp intensity changes and to find object boundaries in an image. The Canny edge detector classifies a pixel as an edge if the gradient magnitude of the pixel is larger than those of pixels at both its sides in the direction of maximum intensity change. In this paper we will show that defining edges in this manner causes some obvious edges to be missed. We will also show how to revise the Canny edge detector to improve its detection accuracy.

Piecewise linear mapping functions for image registration

Abstract A new approach to determination of mapping functions for registration of digital images is presented. Given the coordinates of corresponding control points in two images of the same scene, first the images are divided into triangular regions by triangulating the control points. Then a linear mapping function is obtained by registering each pair of corresponding triangular regions in the images. The overall mapping function is then obtained by piecing together the linear mapping functions.

Registration of images with geometric distortions

A technique for registration of images with geometric distortions is described. This technique uses two surface splines to represent the X-component and the Y-component of a mapping function. A mapping function is described in such a way that it would map corresponding control points in the image exactly on top of each other and map other points in the image by interpolation using information and local geometric distortion between the images. >

Image registration by local approximation methods

Abstract Image registration is approached as an approximation problem. Two locally sensitive transformation functions are proposed for image registration. These transformation functions are obtained by the weighted least-squares method and the local weighted mean method. The former is a global method and uses information about all control points to establish correspondence between local areas in the images; nearby control points are, however, given higher weights to make the process locally sensitive. The latter is a local method and uses information about local control points only to register local areas in the images.

Segmentation of skin cancer images

An automatic method for segmentation of images of skin cancer and other pigmented lesions is presented. This method first reduces a color image into an intensity image and approximately segments the image by intensity thresholding. Then, it refines the segmentation using image edges. Double thresholding is used to focus on an image area where a lesion boundary potentially exists. Image edges are then used to localize the boundary in that area. A closed elastic curve is fitted to the initial boundary, and is locally shrunk or expanded to approximate edges in its neighborhood in the area of focus. Segmentation results from 20 randomly selected images show an average error that is about the same as that obtained by four experts manually segmenting the images.

Fusion of multi-exposure images

A method for fusing multi-exposure images of a static scene taken by a stationary camera into an image with maximum information content is introduced. The method partitions the image domain into uniform blocks and for each block selects the image that contains the most information within that block. The selected images are then blended together using monotonically decreasing blending functions that are centered at the blocks and have a sum of 1 everywhere in the image domain. The optimal block size and width of the blending functions are determined using a gradient-ascent algorithm to maximize information content in the fused image.

A Region-Based Approach to Digital Image Registration with Subpixel Accuracy

Automatic registration of images with translational, rotational, and scaling differences is discussed. To register two images from the same scene, first, the images are segmented and closedboundary regions in the images are extracted. Next, centers of gravity of closed-boundary regions are taken as control points and correspondence is established between the control points. Using this correspondence, the original images are then revisited and the segmentation process is refined in such a way that the obtained corresponding regions become optimally similar. This enables determination of centers of gravity of the regions up to subpixel accuracy. Finally, registration parameters are determined by the least squares error criterion.

A Two-Stage Cross Correlation Approach to Template Matching

Two-stage template matching with sum of absolute differences as the similarity measure has been developed by Vanderburg and Rosenfeld [1], [2]. This correspondence shows the development of two-stage template matching with cross correlation as the similarity measure. The threshold value of the first-stage is derived analytically and its validity is verified experimentally. Considerable speed-up over the one-stage process can be obtained by introducing only a small false dismissal probability.

A comparative study of transformation functions for nonrigid image registration

Transformation functions play a major role in nonrigid image registration. In this paper, the characteristics of thin-plate spline (TPS), multiquadric (MQ), piecewise linear (PL), and weighted mean (WM) transformations are explored and their performances in nonrigid image registration are compared. TPS and MQ are found to be most suitable when the set of control-point correspondences is not large (fewer than a thousand) and variation in spacing between the control points is not large. When spacing between the control points varies greatly, PL is found to produce a more accurate registration than TPS and MQ. When a very large set of control points is given and the control points contain positional inaccuracies, WM is preferred over TPS, MQ, and PL because it uses an averaging process that smoothes the noise and does not require the solution of a very large system of equations. Use of transformation functions in the detection of incorrect correspondences is also discussed.