Enhancing spatial resolution of images from light field imaging systems using sub-pixel disparity

摘要

A light field imaging system captures different images provide views that are shifted by sub-pixel amounts relative to one another. These views can be combined to produce a higher resolution digital image of the object.

主权项

1. A light field imaging system for producing an image of an object, the light field imaging system comprising:
a primary optical subsystem that forms an optical image of an object, the optical image located at an image plane of the primary optical subsystem; a light field sensor module comprising a secondary imaging array and a sensor array, the secondary imaging array imaging a pupil of the primary optical subsystem onto the sensor array, the secondary imaging array displaced from the image plane so that the sensor array simultaneously captures an array of views of the object; an auto defocus module that adjusts a spacing of the light field sensor module relative to the primary optical subsystem, wherein different spacings result in different disparities for the array of views; and a processing module coupled to receive the array of views; wherein the auto defocus module receives feedback from the processing module and uses the feedback to adjust the spacing of the light field sensor module relative to the primary optical subsystem to achieve a view-to-view disparity of M/N pixels, wherein M and N are relatively prime integers and N/M is greater than 1; and the processing module combines the array of views into a digital image of the object having a resolution that is approximately L times higher than a resolution of any one of the views where L is an integer factor of N and L is 2 or greater.