Marmelator/cv-analysis-service
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 37 Wiki Activity

seperate function which is filtering for isolated boxes

Browse Source
This commit is contained in:
llocarnini 2022-01-27 00:19:39 +01:00
parent cf5851b652
commit edf3bfe446
1 changed files with 35 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
table_parsing/table_parsig.py
View File

@ -4,11 +4,11 @@ import cv2
import numpy as np import numpy as np
import pdf2image import pdf2image
from matplotlib import pyplot as plt from matplotlib import pyplot as plt
from timeit import timeit
def parse(image: np.array): def parse(image: np.array):
gray_scale = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) gray_scale = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
th1, img_bin = cv2.threshold(gray_scale, 150, 225, cv2.THRESH_BINARY) th1, img_bin = cv2.threshold(gray_scale, 200, 255, cv2.THRESH_BINARY)
img_bin = ~img_bin img_bin = ~img_bin
line_min_width = 5 line_min_width = 5
@ -25,12 +25,39 @@ def parse(image: np.array):
return labels, stats return labels, stats
# def filter_unconnected_cells(stats):
# filtered_cells = []
# for i, val in enumerate(stats[2:]):
# x, y, w, h, area = stats[i][0], stats[i][1], stats[i][2], stats[i][3], stats[i][4]
# if w > 35 and h > 13 and area > 500:
# # print(stats[i])
# if y == stats[i - 1][1] or y == stats[i + 1][1]:
# filtered_cells.append(stats[i])
# return filtered_cells
#
#
#
# def annotate_image(image, stats):
# stats = filter_unconnected_cells(stats)
# for i,val in enumerate(stats):
# x, y, w, h, area = stats[i][0], stats[i][1], stats[i][2], stats[i][3], stats[i][4]
# cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 255), 2)
#
# for i, (s, v) in enumerate(zip(["x", "y", "w", "h"], [x, y, w, h])):
# anno = f"{s} = {v}"
# xann = int(x + 5)
# yann = int(y + h - (20 * (i + 1)))
# cv2.putText(image, anno, (xann, yann), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 0, 255), 2)
#
# return image
def annotate_image(image, stats): def annotate_image(image, stats):
print(stats.shape)
for i in range(2, len(stats)): for i in range(2, len(stats)):
x,y,w,h,area = stats[i][0],stats[i][1],stats[i][2],stats[i][3],stats[i][4] x, y, w, h, area = stats[i][0], stats[i][1], stats[i][2], stats[i][3], stats[i][4]
if w > 35 and h > 13 and area > 500: if w > 35 and h > 13 and area > 500:
#print(stats[i]) # print(stats[i])
if y == stats[i-1][1] or y == stats[i+1][1]: if y == stats[i - 1][1] or y == stats[i + 1][1]:
cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 255), 2) cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 255), 2)
for i, (s, v) in enumerate(zip(["x", "y", "w", "h"], [x, y, w, h])): for i, (s, v) in enumerate(zip(["x", "y", "w", "h"], [x, y, w, h])):
@ -48,14 +75,17 @@ def parse_tables_in_pdf(pages):
def annotate_tables_in_pdf(pdf_path, page_index=1): def annotate_tables_in_pdf(pdf_path, page_index=1):
timeit()
page = pdf2image.convert_from_path(pdf_path, first_page=page_index + 1, last_page=page_index + 1)[0] page = pdf2image.convert_from_path(pdf_path, first_page=page_index + 1, last_page=page_index + 1)[0]
page = np.array(page) page = np.array(page)
_, stats = parse(page) _, stats = parse(page)
page = annotate_image(page, stats) page = annotate_image(page, stats)
print(timeit())
fig, ax = plt.subplots(1, 1) fig, ax = plt.subplots(1, 1)
fig.set_size_inches(20, 20) fig.set_size_inches(20, 20)
ax.imshow(page) ax.imshow(page)
plt.show() plt.show()