Equipment

We used the following equipment:

• Digital camera: Nikon D70

Image taken from http://www.nikonusa.com/fileuploads/gallery/D70/D70_gallery.html#

• Spectroradiometer: Photo Research PR-650

• Monochromato and Integrating sphere: Oriel Cornerstone 130

• Tungsten White-light: Illumination Technologies 3900

• Computer Code: MATLAB

• Computer: Dell

• Camera Simulator: Professor Wandell’s Imaging Systems Evaluation Tool (ISET)

• Test Image: Macbeth color chart under tungsten illumination

• Optoliner Device

Procedure Overview

We characterized the Nikon D70 digital camera and measured the image detector attributes needed for successful simulation: spectral response function, read noise, dark current, dark signal non-uniformity (DSNU), and photo response non-uniformity (PRNU).  To complete this characterization, we required a variety of specialized equipment.  For the spectral response, we measure this parameter in a calibrated environment, utilizing a tungsten white-light source (Illumination Technologies 3900), an integrating sphere (Oriel Cornerstone 130), a monochromator (Oriel Cornerstone 130), and a spectroradiometer (Photo Research PR-650).  We illuminate the tungsten light, which casts a broadband light with a blackbody spectrum into the monochromator.  Serving the purpose of filtering the white-light, the monochromator produces a narrowband colorful beam with a one to ten nanometer spectral bandwidth.  Next the photons pass into the integrating sphere.  With the function of uniformizing the spatial distribution of the narrowband light signal, the integrating sphere channels the light into the camera sensor.  We also use the spectroradiometer to characterize and calibrate the radiance of this input light signal. 

Each of our measurement procedures are computer-controlled by a MATLAB code, written by our course instructors, utilizing a combination of ‘get’ and ‘set’ commands.  A central command computer operates this procedure through the use of USB cables linked to each of the equipment listed above.  Hence, through the custom designed MATLAB scripts, the computer can conduct the entire calibration and characterization process automatically.  We use these scripts for the measurement of the PRNU values, in addition to the spectral response.

Computed without the need for a light source, the read noise, dark current and DSNU were measured with the lens cap covering the camera lens.  We photograph a sequence of images at various exposure time settings in this dark setting.  We download these dark images into a computer and use MATLAB scripts that we designed to process the image data to determine the desired values: read noise, dark current, and DSNU.

We use the above procedures to identify the required parameters: the spectral response function, read noise, dark noise, DSNU, and PRNU of the Nikon D70 camera.  With these values, we simulate the Nikon D70 digital camera with the use of the Imaging Systems Evaluation Tool (ISET), developed by Professor Wandell’s lab.  As a finale to our project, we contrast an observed data set from the Nikon D70 with the values predicted by the ISET simulator.  Specifically, we photographed the Macbeth color chart under tungsten illumination with the Nikon D70 and compared these results with expected values from the ISET.

Next: Dark Current | Home