Deep Learning-Based Wildfire Image Detection and Classification Systems for Controlling Biomass

Algorithm 1

Computing the hash-table keys.

	Inputs (1): Initial interpolated version of the LR image.
	(2): -Quantization factor for angle (e.g. 24).
	(3): -Quantization factor for strength (e.g. 3).
	(4): -Quantization factor for coherence (e.g. 3).
	Output
	(1): Hash-table keys per pixel, denoted by , , and .
	Process (i) Compute the image gradients
	(ii) Construct the matrix , and obtain the gradients’ angle , strength , and coherence μk
	(iii) Quantize: , where is the ceiling function