find_clearance.py

import numpy as np
import matplotlib.pyplot as plt
import cv2
import sys

f = open(sys.argv[1],"r")
img = np.loadtxt(f) #read content from input file
f.close()

def auto_canny(image, sigma=0.33):
    # compute the median of the single channel pixel intensities
    v = np.median(image)
    # apply automatic Canny edge detection using the computed median
    lower = int(max(0, (1.0 - sigma) * v))
    upper = int(min(255, (1.0 + sigma) * v))
    edged = cv2.Canny(image, lower, upper)
    # return the edged image
    return edged

def denoise(image,lower,upper):
    drawing = np.zeros((image.shape[0], image.shape[1]), dtype=np.uint8)
    for i in range(image.shape[0]):
        for j in range(image.shape[1]):
            if (image[i,j] < lower):
                drawing[i,j] = 0
            elif(image[i,j] > upper):
                drawing[i,j] = 255
            else:
                drawing[i,j]=image[i,j]
    return drawing

#image preprocessing
drawing = denoise(img,1,4)#set threshold for the depth image        
img1  = auto_canny(cv2.GaussianBlur(drawing, (5, 5), 0))
img1  = cv2.dilate(img1, None, iterations=2)
img1  = cv2.erode(img1, None, iterations=2)
cropped = np.zeros((img.shape[0], img.shape[1]), dtype=np.uint8)
cropped[30:108,60:120]=img1[30:108,60:120]# get the interested region

#drawing box
img2, contours, hierarchy = cv2.findContours(cropped, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
contours = sorted(contours, key = cv2.contourArea, reverse = True)[:1]
screenCnt = None
contours_poly = [None]*len(contours)
boundRect = [None]*len(contours)
for i, c in enumerate(contours):
    contours_poly[i] = cv2.approxPolyDP(c, 3, True)
    boundRect[i] = cv2.boundingRect(contours_poly[i])
    
#print result
left = boundRect[0][0]-60
right = 120-(boundRect[0][0]+boundRect[0][2])
print("left",left*1.5/60) if (left > right) else print("right",right*1.5/60)