Thermal FLIR camera reading inaccuracy

[MadsR]

Current model uses the linear scale of FLIR cameras.

Using RAW camera values and calculating surface temperatures based on object parameters, might be more accurate.

Camera might be affected by a significant portion of the image not being the target.

• Copper base plate, PCR plate fills +80% of the image

Temperature Calculations https://flir.custhelp.com/app/answers/detail/a_id/3321/~/the-measurement-formula

Temperature Linear Mode https://flir.custhelp.com/app/answers/detail/a_id/1021/~/temperature-linear-mode

[MadsR]

Radiometric Measurements (https://www.flir.com/discover/security/radiometric/the-benefits-and-challenges-of-radiometric-thermal-technology/) A radiometric thermal camera measures the temperature of a surface by interpreting the intensity of an infrared signal reaching the camera.

Procedure (https://forums.ni.com/t5/Machine-Vision/FLIR-IR-Camera-A655sc-Greyscale-to-Temperature-Conversion/td-p/4146799)

1) The camera can either give raw 16bit images and the software has to do the conversion itself with 3 steps:

• Subtract the global offset from each pixel and divide by gain (easy!)

• To the result, apply the universal thermography equation, with the object parameters (f** hard and a lot of work!)

• To the result apply equation 2.2.4.8 from the GenICam standard: Temperature (in Kelvin) = B / log(R / S + F), where S is the 16-bit digital signal value. log(x) is the base-e logarithm of the x parameter. (easy again!)

[MadsR]

Conclusion was made to not change camera measurement model, to avoid overcomplicating temperature readings.

Instead focus should be directed to investigating camera (and object) parameters to ensure they have been set to values ideal for water surfaces (opened new issue as #58)