Convert image to heatmap

heatmap.py

#!/usr/bin/env python

import cv2
import numpy as np
import argparse


'''
Create blended heat map with JET colormap 

'''

def create_heatmap(im_map, im_cloud, kernel_size=(5,5),colormap=cv2.COLORMAP_JET,a1=0.5,a2=0.5):
    '''
    img is numpy array
    kernel_size must be odd ie. (5,5)
    '''

    # create blur image, kernel must be an odd number
    im_cloud_blur = cv2.GaussianBlur(im_cloud,kernel_size,0)

    # If you need to invert the black/white data image
    # im_blur = np.invert(im_blur)
    # Convert back to BGR for cv2
    #im_cloud_blur = cv2.cvtColor(im_cloud_blur,cv2.COLOR_GRAY2BGR)

    # Apply colormap
    im_cloud_clr = cv2.applyColorMap(im_cloud_blur, colormap)

    # blend images 50/50
    return (a1*im_map + a2*im_cloud_clr).astype(np.uint8) 



if __name__ == "__main__":
    
    ap = argparse.ArgumentParser()
    ap.add_argument('-m','--map-image',default='map.png',required=True,help="Path to map image")
    ap.add_argument('-c','--cloud-image',default='clouds.png',required=True,help="Path to cloud image (grayscale)")
    ap.add_argument('-s','--save',action='store_true',default=False,help="Save image")

    #ap.add_argument('--cloud-pos',default=(0,0),required=False,help="Position of cloud, relative to map origin")
    args = ap.parse_args()

    im_map = cv2.imread(args.map_image)
    im_cloud =cv2.imread(args.cloud_image)
    # Normalize cloud image?
    im_heatmap = create_heatmap(im_map, im_cloud, a1=.5, a2=.5)
    if args.save:
        cv2.imwrite('clouds_map_out.png',im_heatmap)
    else:
        cv2.imshow('cloud cover',im_heatmap)
        cv2.waitKey(0)

Those arguments are required you need to pass them while running the code.
For e.g. python heatmap.py -m map.png -c cloud.png