Typically, if someone wishes to obtain three-dimensional images of micrometer-scale objects, they need to use a device such as a confocal microscope or a white-light interferometer. Such equipment is big, expensive, and often has to be mounted on a vibration-free table. Even then, it can take up to a few hours to get the finished images. Scientists at MIT's Department of Brain and Cognitive Sciences, however, have created a system that can obtain the same kind of images almost instantly, using a soda can-sized sensor and a sheet of rubber. It's called GelSight.

The clear, synthetic GelSight rubber is coated on one side with a layer of paint, that contains tiny flakes of metal. When the painted side of the rubber is pressed against a surface, it deforms to the shape of that surface. Looking through the unpainted side of the rubber, one can see the minute contours of the surface, pressing up into the paint. Using cameras in the sensor along with computer algorithms, the system is able to turn those contours into 3D images, that capture details less than one micrometer in depth and approximately two micrometers in width.

The paint is necessary in order to standardize the optical qualities of the surface. Many surfaces are composed of multiple parts that react to light differently - they could be transparent or opaque, for instance. The paint, however, causes everything that makes up the surface to react to light in the same way. This allows for the use of standard computer vision technologies, for obtaining the images.

A major aerospace company has already expressed an interest in using GelSight to check the structural integrity of its products, while forensics experts are also interested in using the technology to identify the unique marks that individual guns leave on shell casings. The MIT researchers believe it could also have applications in areas such as dermatology and biometrics.

The video below illustrates the kinds of things that the technology is capable of.