firelab-general/ratio_pyrometry.py

import math
import cv2 as cv
import numpy as np
from numba import jit
import json

# Cropping config
# x1 = 420
# x2 = 1200
# y1 = 400
# y2 = -1

# post-processing
smoothing_radius = 2

# temperature key generation
key_entries = 6


@jit(nopython=True)
def rg_ratio_normalize(
    imgarr, 
    I_Darkcurrent,
    f_stop, 
    exposure_time, 
    ISO, 
    MIN_TEMP, 
    MAX_TEMP
):
    # set max & min to most extreme values, 
    # work up & down respectively from there
    tmin = MAX_TEMP
    tmax = 0

    imgnew = imgarr.copy()
    for i in range(len(imgarr)):
        for j in range(len(imgarr[i])):
            px = imgarr[i][j]

            # normalize R & G pixels
            r_norm = (px[0] - I_Darkcurrent) * (f_stop ** 2) / (ISO * exposure_time)
            g_norm = (px[1] - I_Darkcurrent) * (f_stop ** 2) / (ISO * exposure_time)

            # apply camera calibration func
            temp_C = pyrometry_calibration_formula(g_norm, r_norm)

            # remove pixels outside calibration range
            if MAX_TEMP != None and temp_C > MAX_TEMP or MIN_TEMP != None and temp_C < MIN_TEMP:
                temp_C = 0

            # update min & max
            if temp_C < tmin and temp_C >= 0:
                tmin = temp_C
            if temp_C > tmax:
                tmax = temp_C

            imgnew[i][j] = [temp_C, temp_C, temp_C]
    return imgnew, tmin, tmax


@jit(nopython=True)
def pyrometry_calibration_formula(i_ng, i_nr):
    """
    Given the green-red ratio, calculates an approximate temperature 
    in Celsius.
    """
    return 362.73 * math.log10(
        (i_ng/i_nr) ** 3
    ) + 2186.7 * math.log10(
        (i_ng/i_nr) ** 3
    ) + 4466.5 * math.log10(
        (i_ng / i_nr) ** 3
    ) + 3753.5

def ratio_pyrometry_pipeline(
    file_bytes,
    # camera settings
    I_Darkcurrent: float,
    exposure_time: float,
    f_stop: float,
    ISO: float,
    # pyrometry config
    MAX_TEMP: float,
    MIN_TEMP: float
):

    # read image & crop
    img_orig = cv.imdecode(file_bytes, cv.IMREAD_UNCHANGED)
    # img = img[y1:y2, x1:x2]

    img, tmin, tmax = rg_ratio_normalize(
        img_orig,
        I_Darkcurrent,
        f_stop,
        exposure_time,
        ISO,
        MIN_TEMP,
        MAX_TEMP
    )

    # build & apply smoothing conv kernel
    k = []
    for i in range(smoothing_radius):
        k.append([1/(smoothing_radius**2) for i in range(smoothing_radius)])
    kernel = np.array(k)

    img = cv.filter2D(src=img, ddepth=-1, kernel=kernel)

    # write colormapped image
    img_jet = cv.applyColorMap(img, cv.COLORMAP_JET)
    # cv.imwrite(f'{file_name}-cropped-transformed-ratio.{file_ext}', img_jet)

    # --- Generate temperature key ---

    # adjust max & min temps to be the same as the image
    # tmin_adj = tmin / (smoothing_radius ** 2)
    # tmax_adj = tmax / (smoothing_radius ** 2)
    # Generate 6-step key
    step = (tmax - tmin) / (key_entries-1)
    temps = []
    key_img_arr = [[]]
    for i in range(key_entries):
        res_temp = tmin + (i * step)
        res_color = (tmax - (i * step)) / MAX_TEMP * 255
        temps.append(math.floor(res_temp))
        key_img_arr[0].append([res_color, res_color, res_color])

    key_img = np.array(key_img_arr).astype(np.uint8)
    key_img_jet = cv.applyColorMap(key_img, cv.COLORMAP_JET)

    tempkey = {}
    for i in range(len(temps)):
        c = key_img_jet[0][i]
        tempkey[temps[i]] = f"rgb({c[0]}, {c[1]}, {c[2]})"

    # original, transformed, legend
    return img_orig, img_jet, tempkey