When I'm taking photos with the OakD-Pro mono camera (which is sensitive to IR light), the center of the photo is brighter than the edges of the photo. Instead of using the built-in IR floodlight, I'm using evenly spaced IR LED strips (850nm) to illuminate the area. Because of the uniformity in spacing of the external lights, I think I'm seeing the infrared hotspot phenomenon in my photos. I've uploaded two photos with the same external lighting setup, slightly shifted from each other, and both photos are brightest in the center of the image. Can you comment on whether that seems accurate? And if you have any suggestion on how to correct for this (preferably not during post-processing), that would be great!

  • erik replied to this.

    erik Not a great photo, but it's the only one I have at the moment (I can get a better one next week if that would help). There are three rows of IR lights are on the bottom of the metal slats under the black plastic, spaced about 1 inch apart in each row. Adjacent and parallel to them are visible light strips, so there are three rows of lights also spaced about 1 inch apart. The black plastic is to block out the overhead lights from above this setup, and the object to be photographed is placed on the wood below the lights (~1.5m away).

    I was also looking back at a visible light only photo (no IR lighting) I took of the same setup, and it seems to have a similar effect?

    • erik replied to this.

        Hi VivianPatterson , Could you show where LEDs are? It's not that clear.. Also on the visible spectrum, bottom-left looks quite bright, but again I can't say for sure as I am not sure on the source of the light.

            erik You can't see the LEDs in the photo (they're facing down, on the bottom-side of the metal slats), unfortunately! Image the underside of the three metal bars having these LED strips running along them (https://store.waveformlighting.com/products/infrared-850-nm-ir-led-strip-light?src=pse ).

            Here are a couple more photos that might give a better sense of the setup. One is of the IR LED strip before it's unrolled (and I guess this means that my camera does not have the notch filter at 940nm, since these 850nm lights are clearly illuminated here). The second photo is an upward shot of the visible LED strips mounted to the metal bars. The IR strip was mounted right next to it after this photo was taken.

            If this doesn't help clarify, I can get a better photo of the whole setup when I'm back in the office next week. Thank you so much!

            • erik replied to this.

                Hi VivianPatterson ,
                It helps, and I think it is an issue of vignetting, as sensors just weren't tuned for such an environment (wavelength). And as a side note, custom sensor tuning typically costs a few $10k, if you'd be interested in something like this, please send us an email to support@luxonis.com and we can further discuss such custom efforts. Thoughts?

                    erik Thank you so much for your insights! We'll try some 940nm IR lights first and see if that setup works for us.