3D Hyperspectral Visualization and Analysis

Birds and many other organisms see more colors than humans can. Studying avian coloration requires using technologies that have expanded spectral sensitivity including the near UV and visual wavelengths. For nearly two decades, full spectrum measurements have been made using high resolution diode-array spectrophotometers. These are excellent tools, but they only allow for measurement of single tiny patches (~ 3 mm in diameter). To capture an entire avian plumage with full spectral sensitivity would require a non-traditional “bird eye camera” unlike all cameras designed for human vision.      

In this research program, we have developed a next-generation, broadly sensitive (UV-Vis) spectral imaging system for the capture of organismal color reflectance. The system integrates two previously distinct technologies: hyperspectral imaging and 3D laser scanning.  

Hyperspectral imaging involves taking photographs that supply full spectrum information for every pixel in the image. Instead of providing a traditional ‘flat’ photograph with B&W or RGB information at each pixel, a hyperspectral image is a cube with a reflectance spectrum for each pixel.

3D laser scanning involves measuring the reflectance of a laser from an object surface, and producing a computerized grid model of the surface object shape. By placing the object on a turntable and integrating laser scanning data over different positions, we can create a 3D model of the surface of the bird.

In this research project, we have combined these two technologies to create 3D hyperspectral data sets with submillimeter grid resolution for the entire surface of a bird.

These data are analyzed with our new analysis program, BirdViewer, we can convert these surface reflectance spectra into a tetrahedral color vectors, and project these color vectors back on to the surface grid of the specimen. The result is the first integrated, spatially resolved view of full spectrum color variation over the surface of a bird.

The result is a new family of tools for the exploration and analysis of organismal color variation.


Kim, M. K., Harvey, T. A., Kittle, D. S., Rushmeier, H., Dorsey, J., Prum, R. O., and Brady, D. J. 2012. 3D Imaging Spectroscopy for Measuring 3D Hyperspectral Patterns on Solid Objects. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2012) 31: 1-11.