98 lines
3.1 KiB
Python
98 lines
3.1 KiB
Python
import cv2
|
|
import numpy as np
|
|
|
|
|
|
def adaptive_threshold(imagename, invert=False):
|
|
img = cv2.imread(imagename)
|
|
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
|
|
|
|
if invert:
|
|
threshold = cv2.adaptiveThreshold(
|
|
gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY,
|
|
15, -0.2)
|
|
else:
|
|
threshold = cv2.adaptiveThreshold(
|
|
np.invert(gray), 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
|
|
cv2.THRESH_BINARY,
|
|
15, -0.2)
|
|
return img, threshold
|
|
|
|
|
|
def find_lines(threshold, direction=None, scale=15):
|
|
lines = []
|
|
|
|
if direction == 'vertical':
|
|
size = threshold.shape[0] // scale
|
|
el = cv2.getStructuringElement(cv2.MORPH_RECT, (1, size))
|
|
elif direction == 'horizontal':
|
|
size = threshold.shape[1] // scale
|
|
el = cv2.getStructuringElement(cv2.MORPH_RECT, (size, 1))
|
|
elif direction is None:
|
|
raise ValueError("Specify direction as either 'vertical' or"
|
|
" 'horizontal'")
|
|
|
|
threshold = cv2.erode(threshold, el, (-1, -1))
|
|
threshold = cv2.dilate(threshold, el, (-1, -1))
|
|
|
|
dmask = threshold # findContours modifies source image
|
|
|
|
try:
|
|
_, contours, _ = cv2.findContours(
|
|
threshold, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
|
|
except ValueError:
|
|
contours, _ = cv2.findContours(
|
|
threshold, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
|
|
|
|
for c in contours:
|
|
x, y, w, h = cv2.boundingRect(c)
|
|
x1, x2 = x, x + w
|
|
y1, y2 = y, y + h
|
|
if direction == 'vertical':
|
|
lines.append(((x1 + x2) / 2, y2, (x1 + x2) / 2, y1))
|
|
elif direction == 'horizontal':
|
|
lines.append((x1, (y1 + y2) / 2, x2, (y1 + y2) / 2))
|
|
|
|
return dmask, lines
|
|
|
|
|
|
def find_table_contours(vertical, horizontal):
|
|
mask = vertical + horizontal
|
|
|
|
try:
|
|
__, contours, __ = cv2.findContours(
|
|
mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
|
|
except ValueError:
|
|
contours, __ = cv2.findContours(
|
|
mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
|
|
contours = sorted(contours, key=cv2.contourArea, reverse=True)[:10]
|
|
|
|
cont = []
|
|
for c in contours:
|
|
c_poly = cv2.approxPolyDP(c, 3, True)
|
|
x, y, w, h = cv2.boundingRect(c_poly)
|
|
cont.append((x, y, w, h))
|
|
return cont
|
|
|
|
|
|
def find_table_joints(contours, vertical, horizontal):
|
|
joints = np.bitwise_and(vertical, horizontal)
|
|
tables = {}
|
|
for c in contours:
|
|
x, y, w, h = c
|
|
roi = joints[y : y + h, x : x + w]
|
|
try:
|
|
__, jc, __ = cv2.findContours(
|
|
roi, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE)
|
|
except ValueError:
|
|
jc, __ = cv2.findContours(
|
|
roi, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE)
|
|
if len(jc) <= 4: # remove contours with less than 4 joints
|
|
continue
|
|
joint_coords = []
|
|
for j in jc:
|
|
jx, jy, jw, jh = cv2.boundingRect(j)
|
|
c1, c2 = x + (2 * jx + jw) / 2, y + (2 * jy + jh) / 2
|
|
joint_coords.append((c1, c2))
|
|
tables[(x, y + h, x + w, y)] = joint_coords
|
|
|
|
return tables |