From d5c694ac711ca3b434bf16bd920b90d1a7e758c4 Mon Sep 17 00:00:00 2001 From: Samo Penic <samo.penic@gmail.com> Date: Sat, 17 Nov 2018 09:57:31 +0000 Subject: [PATCH] Improving the robustness of all three algorithms. --- aoiOcr.py | 33 ++++++++++++++++++++++++--------- 1 files changed, 24 insertions(+), 9 deletions(-) diff --git a/aoiOcr.py b/aoiOcr.py index 0d59684..bb74503 100644 --- a/aoiOcr.py +++ b/aoiOcr.py @@ -3,14 +3,24 @@ from glob import glob -settings = {"sid_mask": "61xx0xxx", "answer_treshold": 0.25} +settings = {"sid_mask": "64xx0xxx", "answer_treshold": 0.25} classifier = joblib.load("filename.joblib") -#p = Paper(filename="testpage300dpi_scan1.png") -#p=Paper(filename='sizif111.tif', sid_classifier=classifier, settings=settings) +# p = Paper(filename="testpage300dpi_scan1.png") +#p=Paper(filename='sizif111.tif', sid_classifier=classifier, settings={"sid_mask": "11xx0xxx", "answer_treshold": 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) +#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", +] +p=Paper(filename=pa[6], sid_classifier=classifier, settings=settings) # print(p.QRData) # print(p.errors) @@ -23,8 +33,13 @@ print(p.get_paper_ocr_data()) -exit(0) -filelist = glob("processed_scans/*.tif") -for f in filelist: - print(f,Paper(filename=f, sid_classifier=classifier, settings=settings).get_paper_ocr_data()) +filelist = glob("processed_scans/*.tif") +for f in sorted(filelist): + print("processing: {}".format(f)) + print( + f, + Paper( + filename=f, sid_classifier=classifier, settings=settings + ).get_paper_ocr_data(), + ) -- Gitblit v1.9.3