From fe2c1025b102bbf20c9afbc29eedf7a5f069410c Mon Sep 17 00:00:00 2001
From: Samo Penic <samo.penic@gmail.com>
Date: Sat, 17 Nov 2018 14:11:17 +0000
Subject: [PATCH] Readme requirements.
---
aoiOcr.py | 60 +++++++++++++++++++++++++++++++++++++++++++++++++-----------
1 files changed, 49 insertions(+), 11 deletions(-)
diff --git a/aoiOcr.py b/aoiOcr.py
index 72ed3b9..9188b0a 100644
--- a/aoiOcr.py
+++ b/aoiOcr.py
@@ -1,17 +1,55 @@
from Ocr import Paper
from sklearn.externals import joblib
-classifier = joblib.load('filename.joblib')
+from glob import glob
-#p=Paper(filename='testpage300dpi_scan1.png')
-p=Paper(filename='sizif111.tif', sid_classifier=classifier)
-#p=Paper(filename='processed_scans/20141016095134535_0028.tif')
+settings = {"sid_mask": "64xx0xxx", "answer_threshold": 0.25}
+classifier = joblib.load("filename.joblib")
-print(p.QRData)
-print(p.errors)
+#p = Paper(filename="testpage300dpi_scan1.png")
+#p=Paper(filename='sizif111.tif', sid_classifier=classifier, settings={"sid_mask": "11xx0xxx", "answer_threshold": 0.25})
+#p=Paper(filename='processed_scans/20141016095134535_0006.tif', sid_classifier=classifier, settings=settings)
+#p = Paper(filename="processed_scans/20151111080408825_0001.tif",sid_classifier=classifier,settings=settings,)
+#p=Paper(filename='processed_scans/20151028145444607_0028.tif', sid_classifier=classifier, settings=settings)
+pa = [
+ "processed_scans/20141016095134535_0006.tif",
+ "processed_scans/20141016095134535_0028.tif",
+ "processed_scans/20141016095134535_0028.tif",
+ "processed_scans/20141016095134535_0037.tif",
+ "processed_scans/20141021095744144_0005.tif",
+ "processed_scans/20141021095744144_0009.tif",
+ "processed_scans/20141028095553745_0018.tif",
+ "processed_scans/20151013180545275_0011.tif",
+ "processed_scans/20160408140801098_0004.tif",
+ "processed_scans/20160510075445995_0026.tif"
+]
+p=Paper(filename=pa[9], sid_classifier=classifier, settings=settings)
-print(p.getSkewAngle())
-print(p.locateUpMarkers())
-print(p.locateRightMarkers())
-print(p.answerMatrix)
-p.get_enhanced_sid()
+# print(p.QRData)
+# print(p.errors)
+
+# print(p.getSkewAngle())
+# print(p.locateUpMarkers())
+# print(p.locateRightMarkers())
+# print(p.answerMatrix)
+# p.get_enhanced_sid()
+
+
+print(p.get_paper_ocr_data())
+
+
+filelist = glob("processed_scans/*.tif")
+wrong_sid=0;
+total=0
+for f in sorted(filelist):
+ print("processing: {}".format(f))
+ p=Paper(filename=f, sid_classifier=classifier, settings=settings).get_paper_ocr_data()
+ print(f,p)
+ if(p['page_no']==0):
+ total+=1
+ if(len(p['errors'])!=0):
+ wrong_sid+=1
+ if total%10 == 0:
+ print("Total:{}, wrong SID: {}".format(total,wrong_sid))
+
+print("Total:{}, wrong SID: {}".format(total,wrong_sid))
\ No newline at end of file
--
Gitblit v1.9.3