Canon S110 Infrared performence

Posted by Neil Yu on July 11, 2014 at 7:44pm

Hi, I plan to modify the Canon S110 to capture IR image only by replacing the internal filter with a 720 nm long pass filter. I am wondering if anybody has experience with that and how is the image quality in the air. I have read that the point and shoot camera may have visible light leak from the filter and the small sensor size cause issues in IR sensitivity. I am not sure if I need to upgrade the heavier mirrorless camera as my payload is another concern.

Thanks all!

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –
Follow

Replies

  • Vega Tech Group LLC March 20, 2015 at 7:24am

    The Bayer color sensor in the Canon has 1/4 of the total pixels in Red. A Bayer pattern is two Green, one Red, and One Blue pixel grouped together. 

    The best choice is to use either a monochrome sensor where you have the use of all pixels or a camera with a Foveon sensor, where you can take data only from the Red layer. You can use a filter on the Foveon camera if you want just certain wavelengths.

    Also, I suggest a lens with good light transmission in the NIR. As it is the Canon lens used on the S110 is only really good up to about 700nm, light transmission drops off quickly. As it is, this lens has maybe 80% light transmission, if that?

    The sensor in the S110 is not sensitive, so having as much light as possible get through your lens is important.

  • Jeffrey Warren July 16, 2014 at 8:41am

    There's been some work published at PublicLab.org on the S100 and that series, including an infrared conversion:

    https://groups.google.com/forum/#!topic/plots-infrared/9ciA_TB-E-o

    Also some notes on camera filter conversion ease here: http://publiclab.org/wiki/infragram-convertible-cameras

  • Andreas July 15, 2014 at 3:03pm

    http://agribotix.com/blog/

    This guys have awesome information about PA with drones.

  • Vassilis Fotinopoulos July 12, 2014 at 4:22am

    You better have to start with the 680filter, which has more visible details, and after that go to 710-720.

    Pre-consider that sunlight values should be high enough, at least 20.000lux so as to have a good infrared image

    • Neil Yu > Vassilis Fotinopoulos July 12, 2014 at 11:52am

      Hi, Vassilis,

      Thanks for your reply! We are using dual camera system, which includes a unmodified camera to record visible details and the NIR camera. We are interested to use the information from both cameras to estimate NDVI. In your opinion, which filter is more suitable for Canon S110 for this mission, 720 nm or 850 nm, int terms of image quality, IR sensitivity, IR purity, etc..? Thanks again!

    • Houston Howlett > Neil Yu March 13, 2015 at 9:18am

      Neil,

      Can you comment on what platform you are using for the dual camera system? I feel that there is a lot of value in being able to have a visual image as well as an NDVI calculated image when scouting but I would like to capture this in one flight instead of two.

    • Vassilis Fotinopoulos > Neil Yu July 12, 2014 at 12:26pm

      I think this depends exactly on the species you want to observe and the environment you want to observe these species. For that reason i asked you "where on earth are you now"!!!!

    • Neil Yu > Vassilis Fotinopoulos July 12, 2014 at 12:27pm

      I am going to look at the soybean field in solar noon in the summer. North America.

This reply was deleted.