Thermographic data

[sergiud]

Thanks for the dataset. Would it be possible provide the original thermographic data?

While one can work with 8-bit images, their quantization and compression using JPEG are possible sources of bias. Hence, it's rather disadvantageous to use thermographic images that are converted to grayscale images due to loss of precision.

[nixpulvis]

I believe using a BMP format would allow the data to be preserved (lossless) without sacrificing the ability to read the images using standard tools.

[ChrisDAT20]

Thanks for the dataset. Would it be possible provide the original thermographic data?

While one can work with 8-bit images, their quantization and compression using JPEG are possible sources of bias. Hence, it's rather disadvantageous to use thermographic images that are converted to grayscale images due to loss of precision.

Does this mean the dataset is not usable?

[sergiud]

It depends on your notion of "not usable." The combination of quantization and compression along the extremely low resolution (each cell of a standard 6×10 module imaged at 24×40 pixels covers only 4×4 pixels) is certainly not advantageous.

[ChrisDAT20]

I wanted to use this dataset to train a classifier, and then use the classifier with own IR photos. But we have no description which steps have been used to preprocess this photos, nor about the conditions these photos have been taken. If we cannot preprocess our own IR photos the same way it will create even more troubles than the low resolution dataset already has. I contacted the authors of the dataset, but did not get a response.