| commit | author | age | ||
| 0d97e9 | 1 | from aoi_ocr.Ocr import Paper |
| 02e0f7 | 2 | from sklearn.externals import joblib |
| 0d97e9 | 3 | import pkg_resources |
| 5460bf | 4 | |
| SP | 5 | path = "/filename.joblib" # always use slash |
| 6 | filepath = pkg_resources.resource_filename("aoi_ocr", path) | |
| e555c0 | 7 | |
| e2fa6a | 8 | from glob import glob |
| 762a5e | 9 | |
| 9c222b | 10 | settings = {"sid_mask": "11x0xxxx", "answer_threshold": 0.25} |
| 0d97e9 | 11 | classifier = joblib.load(filepath) |
| e555c0 | 12 | |
| 5460bf | 13 | # p = Paper(filename="testpage300dpi_scan1.png") |
| SP | 14 | # p=Paper(filename='sizif111.tif', sid_classifier=classifier, settings={"sid_mask": "11xx0xxx", "answer_threshold": 0.25}) |
| 15 | # p=Paper(filename='processed_scans/20141016095134535_0006.tif', sid_classifier=classifier, settings=settings) | |
| 16 | # p = Paper(filename="processed_scans/20151111080408825_0001.tif",sid_classifier=classifier,settings=settings,) | |
| 17 | # p=Paper(filename='processed_scans/20151028145444607_0028.tif', sid_classifier=classifier, settings=settings) | |
| d5c694 | 18 | pa = [ |
| SP | 19 | "processed_scans/20141016095134535_0006.tif", |
| 20 | "processed_scans/20141016095134535_0028.tif", | |
| 21 | "processed_scans/20141016095134535_0028.tif", | |
| 22 | "processed_scans/20141016095134535_0037.tif", | |
| 23 | "processed_scans/20141021095744144_0005.tif", | |
| 24 | "processed_scans/20141021095744144_0009.tif", | |
| 25 | "processed_scans/20141028095553745_0018.tif", | |
| cf921b | 26 | "processed_scans/20151013180545275_0011.tif", |
| e0996e | 27 | "processed_scans/20160408140801098_0004.tif", |
| 5460bf | 28 | "processed_scans/20160510075445995_0026.tif", |
| d5c694 | 29 | ] |
| 5460bf | 30 | # p=Paper(filename=pa[9], sid_classifier=classifier, settings=settings) |
| bcebb8 | 31 | p = Paper(filename="sizif-test000.tif", sid_classifier=classifier, settings=settings) |
| 02e0f7 | 32 | |
| 0436f6 | 33 | # print(p.QRData) |
| SP | 34 | # print(p.errors) |
| e555c0 | 35 | |
| 0436f6 | 36 | # print(p.getSkewAngle()) |
| 5460bf | 37 | # print(p.locateUpMarkers())%% |
| 0436f6 | 38 | # print(p.locateRightMarkers()) |
| SP | 39 | # print(p.answerMatrix) |
| 40 | # p.get_enhanced_sid() | |
| 41 | ||
| 42 | ||
| 43 | print(p.get_paper_ocr_data()) | |
| e0996e | 44 | |
| d88ce4 | 45 | exit(0) |
| d5c694 | 46 | filelist = glob("processed_scans/*.tif") |
| 5460bf | 47 | wrong_sid = 0 |
| SP | 48 | total = 0 |
| d5c694 | 49 | for f in sorted(filelist): |
| SP | 50 | print("processing: {}".format(f)) |
| 5460bf | 51 | p = Paper( |
| SP | 52 | filename=f, sid_classifier=classifier, settings=settings |
| 53 | ).get_paper_ocr_data() | |
| 54 | print(f, p) | |
| 55 | if p["page_no"] == 0: | |
| 56 | total += 1 | |
| 57 | if len(p["errors"]) != 0: | |
| 58 | wrong_sid += 1 | |
| 59 | if total % 10 == 0: | |
| 60 | print("Total:{}, wrong SID: {}".format(total, wrong_sid)) | |
| e0996e | 61 | |
| 5460bf | 62 | print("Total:{}, wrong SID: {}".format(total, wrong_sid)) |
