Ethan and IPVM, great job as always. It may be of interest that due to the spectral bands being different from VIS to Near IR, the spatial resolution of the image formed changes proportionately. So if the mean wavelength in the VIS is about 550 nm, and the mean for Near IR is about 1100 nm, then the Near IR image will have half the spatial resolution (since the aperture remains constant). But then also the depth of focus shifts with the longer wavelengths forcing the spot on the imager further back. These two factors conspire to make the Near IR images look relatively blurry.
Testing Day vs Night With IR
Published Dec 09, 2014 05:00 AM
** ************ *****.
**** ** **** ****, ********** **** dark, **** ** ****** ********* ********, ****** much ***** ************ ****** (**** '*****' ****) are ***** ** ******* ************* ******** images (***: ***** *** ***** ** ********** ** Shootout).
******* ** ****, ** *** ** surveillance ********* *********** *** **** ******.
*******, *** **** ** **?
************, ** ****** ** ****:
- *** **** * ***** **** ** illumination ** ***** ******* ** *** same ***** ****** *** ***? ***** IR ************ ******* *** ***** **** details (**** *****) ** *** ****** during *** ***? ** ***** ** be ******** ***, ** **, ** how ****?
- **** ******* **** ********* ********** **** using **? *** **** ****** (** not) ** *** ** *** ** 1MP ** *** ***, **** ***** IR?
******* **** ** ** ***** ***** in ******* *** **** *** *** IR ***** *****. ** ** ***** ***** quality *** ************ ***** **** *** time *****, **** **** ***** ** illumination **********. *******, ** *** ***** significantly ****** **** ***-** **** ******.
**** ********** **** *** ********* ******* as **** ******** **** ** ** as ** **** ****** *** ***.
*********** ********** ** ***: **** **** IR, ** *** ****** ** *** the **** ******* ** ***** **** you ** ****** *** ***. ** particular, ** *** ************** *** / ***** *******, ****** **** **.
********, ** ***** / **** ****** are *********, **** *** ** * reason ** ** *** ***** ********** cameras (*** ** ***** ** *****) or ** *** ** *** *** use * ***** ********** ** ***** (either ******* ********** ********** ** * ********).
Image ********
**** ** *** ******* ** * 5MP ******, ***** ** ****** ****** on / ****** *** **** *** full *** *** **** ** ***.
*** ***** ******* ** **** *** day ***** ** ****** ******* ****** than *** ***** *****, ******* ** being **.
**** ** * *** ****** **** the **** **** ******* *******:
**** *** *****, * ******* ******* dropoff ******* *** *** *****.
***, **** ** * ******* ******** between ***, ***** **** ** ** and ***** **** ** ***:
** **** ********** *****, ** **** a '*****' *** ***** ****** ** 0.1 *** ('****** ***' ********). *********, how **** * *** ** ****** will ** ******* **** ** *** specific ***** ***** (***** ***** ** 3 *** ** **** ***** ******* of *.* ***) *** *** *********** of **** ****** (***** *** **** a **** *** ***** ****** ** a **** * ****** ****). *******, this **** ******** * ****** ******* of ** ** ***** ***** ****** than ***** ******* **, *** ***** worse **** *** **** *****.
Other ** *******
***** ******** **** ** ** * 'best ****' / ******** **** ********, where *** ** ** ****** ***** and ** ****.
******* ***** ** ***** **:
- ******* *** ****** *** ***** ** what *** ** *********** *** *****. Recall **** ************ ***** ***** *** fast (******* ****** ****), ** ** *** ****** ** IR ******** *** ****** *****, *** image **** ****** ****** ** **** of * *** ** ******.
- ******* *** *** ***** ** *** camera / ** ******. **** ********** IR ******* *** ****** **** ** detecting *** ************* ********** ***** *** not *** (*** ** ******** ** illuminator) ** ***** ** * **** of ************.
- ** ******** ***** ** *** ******* up **** ****** ***. ** *** IR ********* ** ******** **** *** FoV, ******* ******** ** *** ****** can ***** ** ****.
*** ** **** ****** ** ******** in **** ***** ** ** ********** whether ** *** *********** ********.
Test *********
********** ****** ** **** ****** **** a ****** ****** ** * ****, at ******* ***********, **** ***** ** view *** ***** ******** ********* ****** throughout *** ****. **** ****** ********* ********* within **** ****** ****** ** ******** FOV *** ***** ******. ************, *********** variances ******* ********* *************' ******* *** to ********** **** ** ******, ***, etc. *** *******.
******* **** ****** ******* ** * ~25' ********** ***** ** **** ** order ** **** *** ********** **** (~76 ** *****/~** ** ****) *** fine *********** **********. *** ******** ** this **** **** ***** **** ************ from **** *****, *** *********** ** *** shootout ** *****'* *** ***** ** ******.
*******, ******* **** ****** ***** ******* settings, **** ******** ************ ** */*** and ************ *** ** ~**. ***** different *************' ******* **** **** ** testing. ***** ****** **** ***** *** lux ** **** ***** *** ~*.** lux **** ****** *** *** ** on.
Comments (3)
Skip Cusack
Dec 09, 2014Ethan Ace
Dec 10, 2014Hi Skip, regarding focus shift due to IR light: for this test we refocused all cameras (aside from one, which was fixed focal length) after switching to night with IR on. I actually double checked focus in some cases after switching resolution just to make sure nothing had shifted.
The motorized zoom cameras we used all had auto-focus and could be set to refocus when switching from day to night, but I prefer doing it manually during testing. Sometimes the refocus doesn't go quite right and I want to be sure focus isn't an issue.
UM
Undisclosed Manufacturer #1
Dec 10, 2014One thing that I found interesting while doing tests with IR bullet cameras is what happens to “colors” when switching on and off the IR. On one of the test boards we used a ColorChecker that we printed out at the office, and it gets interesting when you have a decent BW picture without using IR and then switch on the IR. Some of the “colors” completely disappear due to the fact that the IR light is not reflecting the colors from the printer and you get blank spots were you previously at least had something greyish.
Magic :-)