Synopsis

This poster shows results from our development of an extended MATLAB image processing toolbox, which implements some useful optical image processing and mosaicking algorithms found in the literature. We surveyed and selected algorithms from the field which showed promise in application to the underwater environment. We then extended these algorithms to explicitly deal with the unique constraints of underwater imagery in the building of our toolbox. As such, the algorithms implemented include:

1. Contrast limited adaptive histogram specification (CLAHS) to deal with the inherent nonuniform lighting in underwater imagery

2. Fourier based methods for scale, rotation, and translation recovery which provide robustness against dissimilar image regions

3. Local normalized correlation for image registration to handle the unstructured environment of the seafloor

4. Multiresolution pyramidal blending of images to form a composite seamless mosaic without blurring or loss of detail near image borders

Keeping in theme with the global view of CenSSIS, "Diverse Problems, Similar Solutions," many of the algorithms are useful to the rest of the CenSSIS community. Take a look at the normalized correlation section of the poster to see some recent applications of our algorithm to medical imaging.

	Click above to see the figures referenced in the article.

Contrast Limited Adaptive Histogram Specification

The propagation of light underwater suffers from rapid attenuation and extreme scattering. These, in combination with the limited camera-to-light separation available on most underwater imaging platforms, places severe limitations on underwater imagery. To deal with the lighting artifacts of nonuniform illumination and low contrast underwater imagery, we utilize the classical techniques associated with contrast limited adaptive histogram equalization (CLAHE) (Zuiderveld 1994). With this technique the image is broken up into sub-regions. The optimal gray scale distribution is calculated for each of these sub-regions, based upon its histogram and a previously determined transfer function, which is based upon the desired histogram of the sub-region. Then, each pixel of the image is adjusted based upon interpolation between the manipulated histograms of the neighboring sub-regions. Our extensive work upon underwater imagery has suggested that the model of a Raleigh distribution is most suited for underwater imagery.

Fourier Based Image Translation, Scale, and Rotation Recovery

Many image processing problems involve the fundamental task of registration of a pair of images. Methods range from: 1) correlation methods which use pixel values directly; 2) fast Fourier transform methods which use frequency domain information; and 3) feature based methods which use low-level features such as edges and corners. This particular algorithm is based upon Fourier domain methods for scale, rotation, and translation recovery by making use of the phase shift property of Fourier transforms (Reddy 1996).

Local Normalized Correlation

Normalized correlation is a practical measure of similarity (Brown, 1992). Normalized correlation of two signals is invariant to local changes in mean and contrast. When two signals are linearly related, their normalized correlation is 1. When the two signals are not linearly related, but do contain similar spatial variations, normalized correlation will still yield a value close to unity (Irani, 1996).

The lack of rich features in underwater imagery precludes indirect feature based methods, and experimental evidence suggests that direct correlation based methods yield good results. We employ a dense local normalized correlation to determine correspondence between images. The shape of the local normalized correlation surfaces will be concave and have a prominent peak at the correct displacement. We fit a quadratic surface near the surface peak and analytically check for concavity (Mandelbaum, 1999) as a method of outlier rejection.

Multiresolution Pyramidal Based Blending

Due to the rapid attenuation of light underwater, the only way to get a large scale view of the seafloor is to build up a mosaic from smaller local images, such as in Figure 7. The mosaic technique is used to construct an image with a far larger field of view and level of resolution than could be obtained with a single photograph.

Once the mosaic is generated, a technical problem in image representation is joining image borders so that the edge between them is not visible. The two images to be joined may be considered as two surfaces, where the image intensity I(x,y) is viewed as the elevation above the (x,y) plane. The problem then is how to gently distort the images near their common border so that the seam is smooth?

We implement a multiresolution pyramidal blending approach where the two images are decomposed into different band-pass frequency components, merged on those levels, and then reassembled into a single seamless composite image (Burt, 1983). The idea is that with this technique the transition zone between band-pass image components can be appropriately chosen to match the scale of features in that band-pass component.

First, a Gaussian pyramid is constructed for each image where the base level in the pyramid, G₀, is the original image. Each successive level is a low-pass filtered and down-sampled by factor of two version of the previous level for an appropriately chosen kernel w(m,n)). Next, the different band-pass components are formed by generating the Laplacian pyramid. The Laplacian pyramid is generated from the Gaussian pyramid by expanding the image at the next higher level in the pyramid to the resolution of the current level and then subtracting them. This results in each level of the Laplacian pyramid containing a separate band-pass component of the original image. The two Laplacian pyramids are then merged at each level of the pyramid and the resulting new seamless image is constructed from the different pyramid levels via where N is the number of pyramid levels and the notation Ll,l implies expansion of the level Ll,l times to the resolution of G0.

References

Brown, L. G. (1992). "A Survey of Image Registration Techniques." ACM Computing Surveys 24(4): 325376.

Burt, P. J. and E. H. Adelson (1983). "A Multiresolution Spline with Application to Image Mosaics." ACM Transactions of Graphics 2(4): 217236.

Eustice, R., O. Pizarro, et al. (2002). UWIT: Underwater Imaging Toolbox for Optical Image Processing and Mosaicking in MATLAB. Proceedings of the Third Underwater Technology Symposium, 2002, Tokyo, Japan. (to be presented)

Irani, M. and P. Anandan (1996). Robust Multi-Sensor Image Alignment. Sixth International Conference on Computer Vision, 1998.

Mandelbaum, R., G. Salgian, et al. (1999). Correlation-Based Estimation of Ego-Motion and Structure from Motion and Stereo. Proceedings of the Seventh IEEE International Conference on Computer Vision, 1999, Kerkyra, Greece.

Reddy, B. S. and B. N. Chatterji (1996). "An FFT-Based Technique for Translation, Rotation, and Scale-Invariant Image Registration." IEEE Transactions on Image Processing 5(8): 12661271.

Zuiderveld, K. (1994). Contrast Limited Adaptive Histogram Equalization. Graphics Gems IV. P. Heckbert. Boston, Academic Press. IV: 474485.