Hello guys. I’m working on a project to monitor wood chip thermal images and predict its moisture level.
I have an thermal imaging camera but I do not know how to calculate the respective temperature from a given pixel value (RGB). I have a black body so I can set it to a known temperature. Since the pixel values differe as the temperature range of the image, what method should I use to calculate temperature??? Please enlighten me.
do you have any more information about your sensor / camera / driver ? please tell !
do you actually get 16 or 8bit grayscale images ? if you’re getting colored ‘heatmaps’, there’s probably some colour lookup inbetween (unfortunate !).
this pyimgsearch post uses a simple linear approximation, maybe you can start similar, to test & calibrate your real-world hw:
(1) Temperature (°C) = (Gray_16_value)/100 (C) – Absolute_zero (C).
Hii. Thanks for replying. My camera model is AP-TI003 - Non-contact AI Binocular Thermal Imaging Camera Optical Bi-spectrum Fever Detection CCTV IP Camera Scanner from Anpviz, when i asked an SDK from the support team, they sent one. But it shows question marks for the words as it works only in chinese machines (they told when I asked y is that). Also it seems like it is designed to get raw data of the thermal camera, i dont understand. Colour lookup is not available. I want to know if theres a way to identify how the dynamic span changes when the temperature range changes?