Motivation and idea behind this project

I’ve been doing hobby photography for couple decades, starting with old film camera and then quickly going to digital Olympus C4040, E-20N, Nikon D80, D200, D800 and currently enjoying pixels output of Fujifilm GFX system cameras. But all of these cameras designed and optimized to produce images reflecting human eyesight visible colour spectrum. All general purpose cameras implement special filters in front of the actual sensor pixel array surface to remove ultraviolet and infrared light that humans can’t perceive. There is whole science and field of colour designed around rendering of visuals that are compatible with our vision. But light is not limited just to tiny narrow spectrum of colors we see, it expands in both wavelength directions. And CMOS sensors used in majority of cameras these days are sensitive to larger range of wavelength than our eyes.

Based on publications and tests done by others typical CMOS sensor spectral response without additional filters is in range from UV at ~380 nm to NIR up to ~1100 nm. CCD sensors are less sensitive to UV end but more sensitive to NIR up to ~1200 nm. Specialized sensors based on materials like indium-gallium-arsenide can be used to detect and process light with wavelength over 2000 nm. This is still far from thermal cameras which use special bolometer type pixels to detect heat radiation in spectrum range beyound 8000 nm.

But it would be cool to explore NIR and UV photography without specialized equipment. This is a basis of this project today, where I am documenting a modification to my Fujifilm GFX 50S II mirror less larger format camera by removing the visible light bandpass filter element from sensor unit. Doing this unlocks full sensor capability of wider wavelength and will be interesting to experiment in practical photographs. This modification is reversible in case I ever decide to go back, if done carefully.

This camera is not my main body that I am using daily for photography, but an experimental unit that can be risked for this operation. Don’t try this if you are not prepared to deal with possible loss of equipment, if something goes wrong.

All process was documented and recorded with a help of my primary camera that was fixed from water damage some time ago. The IR/UV filter removal process is conceptually same for any type of camera and involve rather deep disassembly of the camera to get access to front of the sensor unit. Interesting to note that Fujifilm itself sells special version of cameras with wide NIR sensitivity. These cameras are significantly more expensive than normal models and available only by special request/customers. Other camera manufacturers also built specialized IR-version cameras for their special customers before.

There are also few recognized 3rd-party vendors who perform conversion of regular camera to wider/specific IR spectrum coverage on commercial basis. LifePixel IR and Kolari Vision are among most well respected ones. But as electronics design engineer it was interesting for me to take camera apart and do the conversion on my own. Also it helps to save precious gold for other projects :)

Teardown and filter removal

Video goes here

Couple of the hi-res photographs of the camera conversion key steps. Fujifilm GFX 50S II camera disassembly process is pretty simple and doable without any special equipment, just a screwdriver and good quality tweezers to pull out various flexible cable connections between modules.

Sensor unit is held in the body by three long bolts. There are fragile shims under each one of them to level the sensor assembly relative to the body. Do not remove them and keep them in place to maintain alignment of the sensor plane!

Then you’d need a soldering iron to desolder ultrasonic actuator power cable from the side of the sensor body and carefully detach the filter with ultrasonic cleaner.

Now sensor is fully exposed to all light. It is also exposed to whatever dust and dirt particles around, so make sure you don’t touch the surface of the sensor and blow off gently any dust particles that lens on sensor protective glass cover. Because of much smaller distance to the pixels even tiny specks of dust will be clearly visible on the photographs, even at fully open apertures like f/4.0 or f/2.8.

FLL perhaps stands for “Filter Lens Layer” for coating above different sub-pixels to provide red, green and blue pattern for colour imaging. MM361A likely stands for Sony IMX361A sensor model that is used in this Fujifilm GFX 50S II camera.

And de-filtered camera is now ready for tests:

Tests and benchmarks

Here are test photographs captured with modified and regular camera during test session October 4, 2026 in Quebec, Canada. All images are corrected for whitebalance and curves with darktable. Most of images captured with Fujifilm GF 120/4 lens and IR pass filter on front.

Regular vision camera (Fujifilm GFX 100 II Project Poseidon) with same lens as reference.

Infrared from modified GFX 50S II with IR filter on GF120/4 lens:

Next photo is one of impressive shots resembling winter, but without actual snow thanks to reflection of green light that turns leaves into white sheets instead. :)

Regular vision camera (Fujifilm GFX 100 II Project Poseidon) with same lens as reference.

Infrared from modified GFX 50S II with IR filter on GF120/4 lens. Interesting that yellow leaves don’t look much different than regular green ones in infrared spectrum.

Next set shows same view but with different spectrums captured.

First is GFX 50S II in infrared with IR pass filter, 1/8” f/11 ISO 100:

Then same modified camera but without any filter on the lens, capturing both UV+visible+IR components. Because IR spectrum contributes much more than other colors we get this sepia look as result, 1/25” f/11 ISO 100:

And finally regular camera (Fujifilm GFX 100 II Project Poseidon) with same lens but with IR passfilter attached, and at much longer exposure 0.8” f/4 ISO 1600. Lot of color is uneven and broken out, compared to modified GFX 50S II above.

Same comparison with tree scenery much closer:

Regular camera (Fujifilm GFX 100 II Project Poseidon) with GF32-64/4 lens, 1/5” f/8 ISO 100

Then modified GFX 50S II body but without any filter on the lens, capturing both UV+visible+IR components. Because IR spectrum contributes much more than other colors we get this sepia look as result, 1/13” f/8 ISO 100:

Capture with GFX 50S II in infrared with IR pass filter with GF32-64/4 lens, 1/8” f/8 ISO 100

GFX 100 mainboard PCB in infrared does not reveal anything too interesting, compared to regular photo from my teardown here

Infrared panoramic image:

Similar panorama with regular vision camera GFX 100 II:

Electric oven lights up nicely:

Night sky also cooperated on the day of conversion and I’ve setup camera with tripod to capture some of the October night sky in Ontario farmland.

	Parameter
Camera	GFX 50S II with filter removed
Lens	Fujifilm GF32-64/4 @ f/4.0
ISO	ISO500
Exposure	25 seconds
Focal	32 mm
File	RAW with minor processing in darktable

I could clearly see purple glow coming out of left bottom corner. Some cameras have some issues related to infrared light leakage or processing, so perhaps 50S II is one of them too? LifePixel IR has interesting note about Fujifilm X-T1 about internal IR contamination after a conversion when the mechanical shutter mode is used.

Once this model is converted to IR, full-spectrum, or H-alpha, the internal IR contamination will affect all images shot at 1 second or more. However, it only shows up with the mechanical shutter. The electronic shutter does not have a problem! Tested at 6,400ISO for 1 second. 

I’ll test this on this converted GFX 50S II too.

Converted camera is also useful for imaging high temperature objects. Here’s operating ERSA I-CON 1 soldering iron with tip temperature +450 °C:

Surface must be above +300 °C to radiate enough infrared heat in near-visible spectrum for such captures to be practical. Pretty cool still to see thermal radiation with such high resolution camera. Same scene recorded in video mode on modified camera:

More captures with IR filter attached to lens

Just like with other set most of images here are with Fujifilm GF 120/4 Macro lens and IR pass filter on front.

Full-spectrum captures without any filter on the lens

Summary and conclusion

And finally, actual modified camera itself with GF 120/4 attached, looking just like any other camera from the outside.

I’ll test more lenses on modified GFX 50S II IR body later on to see how they work in this application. Author of all these photographs also here in infrared during the test shoot session :)

This was a cool weekend project with interesting possibilities for unique beyond visible light photography. I’m looking forward to learn more about infrared photography and have some more UV and IR filters coming soon from epay. I’d also like to give credits to my friend Alex for help and support in making this project possible. Discussion about this article and related stuff is welcome in comment section or at our own IRC chat server: irc.xdevs.com (port 4808, channel: #xDevs.com), via traditional email or on pages of our own xDevs Forum.

Projects like this are born from passion and a desire to share how things work. Education is the foundation of a healthy society - especially important in today's volatile world. xDevs began as a personal project notepad in Kherson, Ukraine back in 2008 and has grown with support of passionate readers just like you. There are no (and never will be) any ads, sponsors or shareholders behind xDevs.com, just a commitment to inspire and help learning. If you are in a position to help others like us, please consider supporting xDevs.com’s home-country Ukraine in its defense of freedom to speak, freedom to live in peace and freedom to choose their way. You can use official site to support Ukraine – United24 or Help99. Every cent counts.