image_reader.py

import numpy as np
import pytesseract
import os, re, sys, cv2
from PIL import Image

def convert_to_grayscale(img_path):
	# Read image with opencv
	img = cv2.imread(img_path)

	# Convert to gray
	img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
	# Apply dilation and erosion to remove some noise
	kernel = np.ones((1, 1), np.uint8)
	img = cv2.dilate(img, kernel, iterations=1)
	img = cv2.erode(img, kernel, iterations=1)

	# Write the image after apply opencv to do some ...
	cv2.imwrite("temp2.png", img)

def overlay_blend(image_path):
	from PIL import Image
	correctionVal = 0.05 # fraction of white to add to the main image
	img_file = Image.open(image_path)
	img_blended = Image.blend(img_file, img_file, correctionVal)

	img_blended.save("temp1.png")

def enhance_image(image_path):
	img = Image.open(image_path)
	width, height = img.size
	new_size = width*6, height*6
	img = img.resize(new_size, Image.LANCZOS)
	img = img.convert('L')
	img = img.point(lambda x: 0 if x < 155 else 255, '1')
	read_text(image_path, img)

def read_text(image_path, image_obj=None):
	img = Image.open(image_path) if not image_obj else image_obj
	imagetext = pytesseract.image_to_string(img)

	output = {}
	re_templates = {
		"Name": ['\Name\s([^\n]+)', '\Name\s([^\s]+)', '\Patient\sName\s([^\n]+)'],
		"Date": ['\Dated[:\-\s]+([^\n|^\s]+)', '\Date[:\-\s]+([^\n|^\s]+)'],
		"Gender": ['\Gender[:\-\s]+([^\n|^\s]+)'],
		"Age": ['\Age[:\-\s]+([^\n|^\s]+)'],
		"Haemoglobin hb, blood": ['haemoglobin[(hb)(blood)\s]+([^\n]+)'],
		"Serum Creatinine": ['serum|creatinine[\s]+([^\n]+)']
	}

	for key in re_templates:
		for match in re_templates[key]:
			temp = filter(None, re.findall("(?i)"+match, imagetext))
			if temp: break

		if temp:
			output[key] = temp[0]

	print(output)

if __name__ == '__main__':
	from sys import argv
	import argparse

	# construct the argument parse and parse the arguments
	ap = argparse.ArgumentParser()
	ap.add_argument("-i", "--image", required=True,
		help="path to input image to be OCR'd")
	ap.add_argument("-p", "--preprocess", type=str, default="thresh",
		help="type of preprocessing to be done")
	args = vars(ap.parse_args())

	if args["preprocess"]=="blend":
		overlay_blend(args["image"])
		read_text("temp1.png")
	elif args["preprocess"]=="grayscale":
		convert_to_grayscale(args["image"])
		read_text("temp2.png")
	elif args["preprocess"]=="enhance":
		enhance_image(args["image"])
	elif args["preprocess"]=="grayscale_enhance":
		convert_to_grayscale(args["image"])
		enhance_image("temp2.png")
	elif args["preprocess"]=="all":
		overlay_blend(args["image"])
		convert_to_grayscale("temp1.png")
		enhance_image("temp2.png")
	else:
		read_text(args["image"])

	# remove temporary files
	for i in ["temp1.png", "temp2.png"]:
		try:
			os.remove(i)
		except Exception:
			continue

	print('\n------ Done -------')