Prev: Simulink: Algebraic loop with frame-based signal
Next: ??? Conversion to cell from double is not possible.???
From: Vinod on 1 Mar 2010 05:49 Hello Friends, How do I calculate the radiance of an image? What exactly is the radiance of an image. Regards, Vinod Karuvat.
From: us on 1 Mar 2010 06:01 "Vinod " <vinodkaruvat(a)gmail.com> wrote in message <hmg631$jk3$1(a)fred.mathworks.com>... > Hello Friends, > > How do I calculate the radiance of an image? What exactly is the radiance of an image. > > Regards, > Vinod Karuvat. a hint: http://dspace.mit.edu/bitstream/handle/1721.1/6878/AITR-623.pdf?sequence=2 us
From: ImageAnalyst on 1 Mar 2010 07:55
Vinod : OK I think we have a terminology problem here. As far as I know you can't have radiance of an image. An image is a recording of light emitted by a scene (either emitted by objects in the scene or light reflected off objects in the scene). The objects in the scene will have radiance but the image will not. There will be "irradiance" on the imaging sensor. Let's review some definitions. From Wikipedia: "Radiance and spectral radiance are radiometric measures that describe the amount of light that passes through or is emitted from a particular area, and falls within a given solid angle in a specified direction. They are used to characterize both emission from diffuse sources and reflection from diffuse surfaces. The SI unit of radiance is watts per steradian per square metre (W·sr-1·m-2). Irradiance, radiant emittance, and radiant exitance are radiometry terms for the power per unit area of electromagnetic radiation at a surface. "Irradiance" is used when the electromagnetic radiation is incident on the surface. "Radiant exitance" or "radiant emittance" is used when the radiation is emerging from the surface. The SI units for all of these quantities are watts per square meter (W/m2), while the cgs units are ergs per square centimeter per second (erg·cm-2·s-1, often used in astronomy). " Let's make a thought experiment. You take a photo of the sun through your telescope and with a very short exposure time and produce a dim image of this very bright object. Then you use the same camera to take a photo of a 100 watt light bulb. Let's say the photo of the light bulb has greater gray levels than your photo of the sun. Now, which object has greater radiance? Answer: the sun, despite it having a darker image. What IS the radiance of the sun and the light bulb? Well to know that you'd have to know a lot about the geometry and optics of the situation (F-stop, aperture diameter, pixel size, etc.), as well as having a calibrated sensor where you know how many gray levels correspond to how many incident watts/square meter. Do you have all of that? Perhaps you're talking about how bright an "image" on a computer monitor is. In this situation they use a unit called a "nit." A nit is a unit of measurement of luminance, or the intensity of visible light, where one nit is equal to one candela per square meter. I'm sure you can find information on the web about how to measure monitor luminance. (Note: luminance is different than radiance). So I doubt you want radiance. Perhaps you want luminance or irradiance, but I'm not sure. Either way, you'll need more info than just gray levels if you want to obtain it in real world SI units. The units of optics, photometry, and radiometry are tricky and confusing so don't feel bad. |