Improved quantitative imaging using synthetic dark frames

Digital Clarity Consultants, Lyons OR

Investigators

Abstract

DESCRIPTION (provided by applicant): Our objective is to demonstrate the removal of dark current from raw images using synthetic dark frames. In CCD and CMOS sensors, dark current varies with sensor temperature and integration time. In many science applications, however, sensor temperature is either not known or not controlled, resulting in inaccurate subtraction of dark frames and greater image noise than necessary. Because dark current is described by a pixel-specific parameter and several global parameters, once these parameters are determined, it is possible to create a synthetic dark frame for any detector temperature or integration time. Our key innovation is a technique for detecting the signature of the dark frame in the raw image, then using the dark-frame signature to deduce the sensor temperature and integration time. .We will deliver demonstration software able to perform robust, accurate dark-frame subtraction from images acquired at different sensor temperatures and integrations times. The overarching goal of DCC is the improvement of science content in images. The largest area where the science content of images is paramount is in biomedicine: applications where our technology can improve the science content ranges from microarrays to breast biopsy technology. Immediate commercial applications include fluorescence microscopy, medical imaging, Raman spectroscopy, and electron microscopy. It is to be noted that this technique is applicable even for old images. On a slightly longer term, i.e., two to three years, applications will be found in high-end consumer digital camera customers. The largest economic impact to the company will be felt here. As an example, the Canon-G6, just announced in August 2004, does old-fashioned dark current correction without possibility for user override! With the use of our product and a routine calibration sequence, users taking images will be assured that dark current noise has been optimally corrected in near real time. The relevance to public health: Modern medical techniques rely heavily on imaging for diagnosis and to develop fundamental knowledge about living systems. Such imaging is hampered by noise from the imaging sensors. Synthetic dark frames offer a cost efficient and automated system for improvement of medical diagnosis and for improvement of the science content in the images used to understand diseases and to develop cures.