add scribble image processing

markup/img.py

@@ -0,0 +1,109 @@
+#from __future__ import absolute_import, unicode_literals
+
+import sys, os.path, re, json, pickle, subprocess
+import shutil
+
+from PIL import Image, ImageFilter
+import numpy
+import imutils
+import cv2
+import dumper
+import random as rng
+from pathlib import Path
+
+from django.conf import settings
+
+WORKDIR = os.path.join(settings.ASSET_DIR, 'markup', 'work')
+
+# https://www.pyimagesearch.com/2014/10/20/finding-shapes-images-using-python-opencv/
+
+
+def find_shapes(image_path):
+    path = Path(image_path)
+
+    img = Image.open(image_path, 'r')
+    if not img.mode in ('RGBA', 'LA'):
+        print('no alpha channel: {}'.format(img.mode))
+        return None
+
+    alpha_layer = img.convert('RGBA').split()[-1]
+    
+    alpha_layer = alpha_layer.filter(ImageFilter.GaussianBlur(5))
+    
+    threshold = 5
+    alpha_layer = alpha_layer.point(lambda p: p > threshold and 255)
+
+    threshold = numpy.array(alpha_layer)
+
+    #bnw = Image.new('RGBA', img.size, (255, 255, 255))
+    #src = cv2.imread(image_path)
+
+    # TODO just get alpha, 1 bit
+    #threshold = cv2.cvtColor(src, cv2.COLOR_BGR2GRAY)
+    #threshold = cv2.blur(threshold, (4,4))
+
+    thresh_path = str(path.with_suffix('.thresh.png'))
+    print('write to', thresh_path)
+    cv2.imwrite(thresh_path, threshold)
+
+    # find all the 'not black' shapes in the image
+    # upper = np.array([255, 255, 255])
+    # lower = np.array([15, 15, 15])
+    # shapeMask = cv2.inRange(image, lower, upper)
+
+    # # alternate to below:
+    # #imgray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+    # #ret, thresh = cv2.threshold(imgray, 127, 255, 0)
+    # #im2, contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
+    # #cv2.imshow('Thresh', thresh)
+    # #cv2.waitKey(0)
+
+    #threshold = 100
+    #canny_output = cv2.Canny(threshold, threshold, threshold * 2)
+    #contours = cv2.findContours(canny_output, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
+    contours = cv2.findContours(threshold, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
+    #print("contours: {}".format(contours))
+
+    # find the contours in the mask
+    #contours = cv2.findContours(shapeMask.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+    #contours = cv2.findContours(shapeMask.copy(), cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
+    contours = imutils.grab_contours(contours)
+    print("{} shapes".format(len(contours)))
+    #cv2.imshow("Mask", shapeMask)
+
+    # Find the convex hull object for each contour
+    hull_list = []
+    #for i in range(len(contours)):
+    for c in contours:
+        hull = cv2.convexHull(c)
+        hull_list.append(hull)
+
+    # Draw contours + hull results
+    #contour_image = shapeMask.copy() # np.zeros((shapeMask.shape[0], shapeMask.shape[1], 3), dtype=numpy.uint8)
+    contour_image = numpy.zeros((threshold.shape[0], threshold.shape[1], 3), dtype=numpy.uint8)
+    for i in range(len(contours)):
+        # compute the center of the contour
+        M = cv2.moments(contours[i])
+        if M["m00"] == 0: M["m00"] = 0.00001
+        cX = int(M["m10"] / M["m00"])
+        cY = int(M["m01"] / M["m00"])
+        color = (rng.randint(0,512), rng.randint(0,512), rng.randint(0,512))
+        cv2.drawContours(contour_image, contours, i, color)
+        #cv2.drawContours(contour_image, hull_list, i, color)
+        x, y, w, h = cv2.boundingRect(contours[i])
+        cv2.rectangle(contour_image, (x,y), (x+w,y+h), color, 1)
+        cv2.circle(contour_image, (cX, cY), 2, color, -1)
+        # cv2.putText(contour_image, "center", (cX - 20, cY - 15),
+        #             cv2.FONT_HERSHEY_SIMPLEX, 0.5, color, 1)
+
+    contour_path = str(path.with_suffix('.contour.png'))
+    print('write to', contour_path)
+    cv2.imwrite(contour_path, contour_image)
+
+    # for c in contours:
+    #     #print json.dumps(c)
+    #     print(dumper.dump(c))
+    #     # draw the contour and show it
+    #     # cv2.drawContours(image, [c], -1, (0, 255, 0), 2)
+    #     # cv2.imshow("Image", image)
+    #     # cv2.waitKey(0)

markup/pdf.py

@@ -130,4 +130,4 @@ def parse_pdf(fname, debug=0):
 
     fp.close()
 
-    return [prod_boxes, scribbles]
+    return [list(filter(None, prod_boxes)), list(filter(None, scribbles))]

markup/work/test_pdf.py

@@ -6,9 +6,8 @@ import inspect
 currentdir = os.path.dirname(os.path.abspath(inspect.getfile(inspect.currentframe())))
 parentdir = os.path.dirname(currentdir)
 parentparentdir = os.path.dirname(parentdir)
-sys.path.insert(0, parentparentdir) 
+sys.path.insert(0, parentparentdir)
 
-import dumper
 import getopt
 import django
 from django.conf import settings
@@ -16,8 +15,7 @@ from django.conf import settings
 os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'procat2.settings')
 django.setup()
 
-from markup.utils import parse_pdf
-from procat2.settings import ASSET_DIR
+from markup.pdf import parse_pdf
 
 
 def main(argv):
@@ -33,6 +31,8 @@ def main(argv):
     for (k, v) in opts:
         if k == '-d': debug += 1
 
-    parse_pdf(args[0], debug)
+    (prods, scribbles) = parse_pdf(args[0], debug)
+    print('prods', scribbles)
+
 
 if __name__ == '__main__': sys.exit(main(sys.argv))

markup/work/test_scribble.py

@@ -0,0 +1,38 @@
+#!/usr/bin/env python3
+import sys
+import os
+import inspect
+
+currentdir = os.path.dirname(os.path.abspath(inspect.getfile(inspect.currentframe())))
+parentdir = os.path.dirname(currentdir)
+parentparentdir = os.path.dirname(parentdir)
+sys.path.insert(0, parentparentdir)
+
+#import dumper
+import getopt
+import django
+from django.conf import settings
+
+os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'procat2.settings')
+django.setup()
+
+from markup.img import find_shapes
+
+
+def main(argv):
+    def usage():
+        print('usage: %s [-d] file ...' % argv[0])
+        return 100
+    try:
+        (opts, args) = getopt.getopt(argv[1:], 'd')
+    except getopt.GetoptError:
+        return usage()
+    if not args: return usage()
+    debug = 0
+    for (k, v) in opts:
+        if k == '-d': debug += 1
+
+    find_shapes(args[0])
+
+
+if __name__ == '__main__': sys.exit(main(sys.argv))

requirements.txt

@@ -15,16 +15,20 @@ django-webpack-loader==0.6.0
 Dumper==1.2.0
 humanize==0.5.1
 importlib-metadata==0.23
+imutils==0.5.3
 ipdb==0.11
 ipython==7.3.0
 ipython-genutils==0.2.0
 jedi==0.13.3
 kombu==4.6.5
 more-itertools==7.2.0
+numpy==1.17.2
+opencv-python==4.1.1.26
 parso==0.3.4
 pdfminer==20191010
 pexpect==4.6.0
 pickleshare==0.7.5
+Pillow==6.2.0
 prompt-toolkit==2.0.9
 psycopg2-binary==2.7.7
 ptyprocess==0.6.0