Pixels Determine Potential, Not Quality
"Pixels Determine Potential, Not Quality" - very well said.
Point well made on pixel potential. Another area you may have already covered but I failed to locate, is the pixel as it stands alone. Maybe you have an opinion on the future of camera specifications and whether manufacturers will need to become more detailed in how they arrive at a product specification based on the construction of the sensor. just a few random thoughts here:
Are all pixels created equal? Isn't pixel size the most important part of imaging? When comparing same sized sensors with the same pixel count, one may have smaller pixels than another which could affect what? Is one sized pixel better than the other smaller or larger? Is this what makes one camera better than another?
If the cameras success is so dependent on 6 factors, how does this information help anyone decipher a manufacturer's specification or brochure to decide between the 100's of choices in cameras from 1MP to 5MP now? ie. what is the best sensor? what is the best SNR? what is the best WDR? What is the best light? What is the best sensitivity range? etc.
David, great observations / questions.
"Are all pixels created equal?"
No, just like all apples are not created equal. Problem #1 is figuring out how they truly differ and how significant that difference is. I just don't know. Also, since most major manufacturers obscure where they get their sensors from (is it an Optina XYZ or an Omnivision ABC), it's hard to even guess.
"Isn't pixel size the most important part of imaging?"
I think pixel size is quite overrated. We are doing a study of pixel size / imager size / resolution comparing it to our test results. My gut feel though is that the correlation between pixel size and quality is weak (even for low light where pixel size is most often argued as a key driver).
"If the cameras success is so dependent on 6 factors, how does this information help anyone decipher a manufacturer's specification or brochure to decide"
I do not think this is possible. Specifications/brochures are good starting points (what are the physical specs - focal length, f stop, pixel count, audio, IR, etc., etc.) but the things that truly differentiate image quality - low light performance, WDR capability, integrated IR adjustment - are hard to determine by simply comparing specs. That's why we do tests and that's why we find variances between specs and performances so commonly.
Is there a specific formula or rule of thumb to figuring out ppf (Example from above: 1mp camera with subject at 50' wide FoV= 25 ppf), or has this been determined from your own research? Im having trouble because it seems there are so many variables that there can't be a uniform way to do this. Perhaps some analyzing software? Thanks
Figuring out how many PPF you need or how many PPF you will get in a given scene? The latter is pretty simple - the formula is just horizontal pixel count / horizontal FoV width.
How many PPF you need is a little more tricky because there are a bunch of things that go into. See our Definitive Guide to PPF for an answer.
Perceived quality per pixel density is sometimes hard to overcome with the general public / business environment. In general, a quick metric is what is wanted in determining if a system is better than another. Unfortunately an attempt to establish a quick metric for a complex system usually means either don't have/want to take the time to understand or is a marketing ploy...
The movie Spinal Tap has a great analogy for an incorrect use of a simple metric for an amplifer, "these go to 11."
Just my rambling thoughts on the matter. Regards.
Ask questions and get answers to your physical security questions from IPVM team members and fellow subscribers.