From 0d97e9b2d738682ed0aa6349b43a9719e0ca0aa9 Mon Sep 17 00:00:00 2001
From: Samo Penic <samo.penic@gmail.com>
Date: Sat, 17 Nov 2018 18:55:57 +0000
Subject: [PATCH] Created package and setup.py.
---
scan2db.py | 8 ++
aoiOcr.py | 7 +
aoi_ocr/filename.joblib | 0
aoi_ocr/template.png | 0
aoi_ocr/__init__.py | 0
setup.py | 12 ++++
aoi_ocr/Ocr.py | 23 ++++---
aoi_ocr/sid_process.py | 27 +++++---
.idea/sonarIssues.xml | 80 ++++++++++++++++++++++++++
aoi_ocr/template-8.png | 0
10 files changed, 135 insertions(+), 22 deletions(-)
diff --git a/.idea/sonarIssues.xml b/.idea/sonarIssues.xml
index 71b5cb9..9f7c463 100644
--- a/.idea/sonarIssues.xml
+++ b/.idea/sonarIssues.xml
@@ -18,6 +18,61 @@
<set />
</value>
</entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/aoi/settings.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/aoi/urls.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/exam/views.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/organization/models.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/organization/urls.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/requirements.txt">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/studentpages/static/studentpages/ux/basic-widgets-sl.js">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/studentpages/static/studentpages/ux/basic-widgets.js">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/studentpages/templates/studentpages/index.html">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/studentpages/urls.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/../../django/sizif-web/aoi/studentpages/views.py">
+ <value>
+ <set />
+ </value>
+ </entry>
<entry key="$PROJECT_DIR$/Ocr.py">
<value>
<set />
@@ -33,11 +88,36 @@
<set />
</value>
</entry>
+ <entry key="$PROJECT_DIR$/scan2db.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/setup.py">
+ <value>
+ <set />
+ </value>
+ </entry>
<entry key="$PROJECT_DIR$/sid_process.py">
<value>
<set />
</value>
</entry>
+ <entry key="$PROJECT_DIR$/sizif-ocr/Ocr.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/sizif_ocr/Ocr.py">
+ <value>
+ <set />
+ </value>
+ </entry>
+ <entry key="$PROJECT_DIR$/sizif_ocr/sid_process.py">
+ <value>
+ <set />
+ </value>
+ </entry>
</map>
</option>
</component>
diff --git a/aoiOcr.py b/aoiOcr.py
index 9188b0a..be3f863 100644
--- a/aoiOcr.py
+++ b/aoiOcr.py
@@ -1,10 +1,13 @@
-from Ocr import Paper
+from aoi_ocr.Ocr import Paper
from sklearn.externals import joblib
+import pkg_resources
+path = '/filename.joblib' # always use slash
+filepath = pkg_resources.resource_filename('aoi_ocr', path)
from glob import glob
settings = {"sid_mask": "64xx0xxx", "answer_threshold": 0.25}
-classifier = joblib.load("filename.joblib")
+classifier = joblib.load(filepath)
#p = Paper(filename="testpage300dpi_scan1.png")
#p=Paper(filename='sizif111.tif', sid_classifier=classifier, settings={"sid_mask": "11xx0xxx", "answer_threshold": 0.25})
diff --git a/Ocr.py b/aoi_ocr/Ocr.py
similarity index 94%
rename from Ocr.py
rename to aoi_ocr/Ocr.py
index 6e9a082..d743ecc 100644
--- a/Ocr.py
+++ b/aoi_ocr/Ocr.py
@@ -1,8 +1,13 @@
from pyzbar.pyzbar import decode
-from sid_process import getSID
+from .sid_process import getSID
import cv2
import numpy as np
-import math
+
+import pkg_resources
+
+markerfile = '/template.png' # always use slash
+markerfilename = pkg_resources.resource_filename(__name__, markerfile)
+
class Paper:
@@ -27,7 +32,7 @@
return
self.imgHeight, self.imgWidth = self.img.shape[0:2]
- def saveImage(self, filename="debug_image.png"):
+ def saveImage(self, filename="/tmp/debug_image.png"):
cv2.imwrite(filename, self.img)
def runOcr(self):
@@ -98,7 +103,7 @@
def getSkewAngle(self):
neg = 255 - self.bwimg # get negative image
- cv2.imwrite("debug_1.png", neg)
+ cv2.imwrite("/tmp/debug_1.png", neg)
angle_counter = 0 # number of angles
angle = 0.0 # collects sum of angles
@@ -123,11 +128,11 @@
except:
skew = 0
- cv2.imwrite("debug_2.png", cimg)
+ cv2.imwrite("/tmp/debug_2.png", cimg)
return skew
def locateUpMarkers(self, threshold=0.85, height=200):
- template = cv2.imread("template.png", 0)
+ template = cv2.imread(markerfilename, 0)
w, h = template.shape[::-1]
crop_img = self.img[0:height, :]
res = cv2.matchTemplate(crop_img, template, cv2.TM_CCOEFF_NORMED)
@@ -158,13 +163,13 @@
for pt in zip(*loc[::-1]):
cv2.rectangle(cimg, pt, (pt[0] + w, pt[1] + h), (0, 255, 255), 2)
- cv2.imwrite("debug_3.png", cimg)
+ cv2.imwrite("/tmp/debug_3.png", cimg)
self.xMarkerLocations = loc
return loc
def locateRightMarkers(self, threshold=0.85, width=200):
- template = cv2.imread("template.png", 0)
+ template = cv2.imread(markerfilename, 0)
w, h = template.shape[::-1]
crop_img = self.img[:, -width:]
res = cv2.matchTemplate(crop_img, template, cv2.TM_CCOEFF_NORMED)
@@ -195,7 +200,7 @@
for pt in zip(*loc[::-1]):
cv2.rectangle(cimg, pt, (pt[0] + w, pt[1] + h), (0, 255, 255), 2)
- cv2.imwrite("debug_4.png", cimg)
+ cv2.imwrite("/tmp/debug_4.png", cimg)
self.yMarkerLocations = [loc[0], loc[1] + self.imgWidth - width]
return self.yMarkerLocations
diff --git a/aoi_ocr/__init__.py b/aoi_ocr/__init__.py
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/aoi_ocr/__init__.py
diff --git a/filename.joblib b/aoi_ocr/filename.joblib
similarity index 100%
rename from filename.joblib
rename to aoi_ocr/filename.joblib
Binary files differ
diff --git a/sid_process.py b/aoi_ocr/sid_process.py
similarity index 91%
rename from sid_process.py
rename to aoi_ocr/sid_process.py
index f21fafb..b14d1d3 100644
--- a/sid_process.py
+++ b/aoi_ocr/sid_process.py
@@ -2,6 +2,11 @@
import numpy as np
from skimage import morphology, img_as_ubyte
+import pkg_resources
+
+templatefile = '/template-8.png' # always use slash
+template8 = pkg_resources.resource_filename(__name__, templatefile)
+
def kernel(x, y):
"""
@@ -24,7 +29,7 @@
# find biggest block of pixels
image1 = cv2.morphologyEx(image2, cv2.MORPH_DILATE, kernel(5, 25), iterations=4)
image1 = img_as_ubyte(image1 > 50)
- cv2.imwrite("sidblock1.png", image1)
+ cv2.imwrite("/tmp/sidblock1.png", image1)
im2, ctrs, hier = cv2.findContours(
image1.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
)
@@ -77,7 +82,7 @@
roi = cv2.resize(roi, (32, 32))
# cv2.rectangle(image,(x,y),( x + w, y + h ),(0,255,0),2)
- cv2.imwrite("sid_no_{}.png".format(i), roi)
+ cv2.imwrite("/tmp/sid_no_{}.png".format(i), roi)
sid_no = sid_no + str(classifier.predict(roi.reshape(1, -1) / 255.0)[0])
return sid_no
@@ -96,14 +101,14 @@
sid_no = ""
sid_len = len(sid_mask)
image=find_biggest_blob(image,original_image,sid_mask)
- cv2.imwrite("sidblock2.png", image)
+ cv2.imwrite("/tmp/sidblock2.png", image)
imgHeight, imgWidth = image.shape[0:2]
numWidth = int(imgWidth / (sid_len))
for i in range(0, sid_len):
num = image[:, i * numWidth : (i + 1) * numWidth]
num = img_as_ubyte(num < 128)
num = cv2.resize(num, (32, 32))
- cv2.imwrite("sid_no_{}.png".format(i), num)
+ cv2.imwrite("/tmp/sid_no_{}.png".format(i), num)
sid_no = sid_no + str(classifier.predict(num.reshape(1, -1) / 255.0)[0])
return sid_no
@@ -123,8 +128,8 @@
original_image, cv2.MORPH_CLOSE, kernel(2, 2), iterations=10
)
block_image = img_as_ubyte(block_image < 50)
- cv2.imwrite("sid_3rd1.png", block_image)
- template = cv2.imread("template-8.png", 0)
+ cv2.imwrite("/tmp/sid_3rd1.png", block_image)
+ template = cv2.imread(template8, 0)
w, h = template.shape[::-1]
res = cv2.matchTemplate(block_image, template, cv2.TM_CCOEFF_NORMED)
loc = np.where(res >= 0.75)
@@ -147,7 +152,7 @@
points = [loc_filtered_y[a], loc_filtered_x[a]]
for pt in zip(*points[::-1]):
cv2.rectangle(cimg, pt, (pt[0] + w, pt[1] + h), (0, 255, 255), 2)
- cv2.imwrite("sid_3rd2.png", cimg)
+ cv2.imwrite("/tmp/sid_3rd2.png", cimg)
sid_no = ""
for i, pt in enumerate(zip(*points[::-1])):
@@ -158,7 +163,7 @@
num = cv2.resize(num, (32, 32))
except:
return ""
- cv2.imwrite("sid_3no_{}.png".format(i), num)
+ cv2.imwrite("/tmp/sid_3no_{}.png".format(i), num)
sid_no = sid_no + str(classifier.predict(num.reshape(1, -1) / 255.0)[0])
return sid_no
@@ -182,7 +187,7 @@
image = 255 - image
image_original = image.copy()
image = img_as_ubyte(image > 100)
- cv2.imwrite("enSID0.png", image)
+ cv2.imwrite("/tmp/enSID0.png", image)
# Remove noise
image = cv2.morphologyEx(image, cv2.MORPH_OPEN, kernel(2, 2), iterations=3)
@@ -197,7 +202,7 @@
# Skeletonization
image = img_as_ubyte(morphology.thin(image > 128))
- cv2.imwrite("enSID1.png", image)
+ cv2.imwrite("/tmp/enSID1.png", image)
# Stub removal (might not be necessary if thinning instead of skeletonize is used above
# Making lines stronger
@@ -207,7 +212,7 @@
# Thining again
image = img_as_ubyte(morphology.skeletonize(image > 0.5))
image = cv2.morphologyEx(image, cv2.MORPH_DILATE, kernel(10, 10))
- cv2.imwrite("enhancedSID.png", image)
+ cv2.imwrite("/tmp/enhancedSID.png", image)
sid_no = segment_by_contours(image, image_original, classifier, sid_mask)
diff --git a/template-8.png b/aoi_ocr/template-8.png
similarity index 100%
rename from template-8.png
rename to aoi_ocr/template-8.png
Binary files differ
diff --git a/template.png b/aoi_ocr/template.png
similarity index 100%
rename from template.png
rename to aoi_ocr/template.png
Binary files differ
diff --git a/scan2db.py b/scan2db.py
new file mode 100644
index 0000000..c150c03
--- /dev/null
+++ b/scan2db.py
@@ -0,0 +1,8 @@
+import os
+import django
+os.chdir('/home/samo/programiranje/django/sizif-web/aoi')
+os.environ.setdefault("DJANGO_SETTINGS_MODULE", "aoi.settings")
+django.setup()
+from exam import models
+
+
diff --git a/setup.py b/setup.py
new file mode 100644
index 0000000..3a0513e
--- /dev/null
+++ b/setup.py
@@ -0,0 +1,12 @@
+#!/usr/bin/python3
+from distutils.core import setup
+
+setup(
+ name="aoi_ocr",
+ version="0.1a",
+ description="OCR of aoi papers.",
+ author="Samo Penic",
+ author_email="samo.penic@gmail.com",
+ url="",
+ packages=["aoi_ocr"],
+)
--
Gitblit v1.9.3