sporada/ocr
1
0
Fork 0
mirror of https://gitlab.ub.uni-bielefeld.de/sfb1288inf/ocr.git synced 2024-12-27 10:44:17 +00:00
Code Issues Projects Releases Wiki Activity

Aktualisieren ocr

Browse Source
This commit is contained in:
Patrick Jentsch 2019-04-11 13:46:24 +02:00
parent fd7ad08e1e
commit d4218fcd7c
1 changed files with 6 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
ocr
View File

@ -120,14 +120,12 @@ class OCRWorkflow(WorkflowRunner):
ocropusnlbin_jobs = [] ocropusnlbin_jobs = []
ocropusnlbin_job_number = 0 ocropusnlbin_job_number = 0
for job in self.jobs: for job in self.jobs:
# This list is empty if you don't wait for ocropus_nlbin_jobs to complete
for file in os.listdir(os.path.join(job["output_dir"], "tmp", "tiff_files")):
ocropusnlbin_job_number += 1 ocropusnlbin_job_number += 1
cmd = 'ocropus-nlbin -o "%s" "%s"' % ( cmd = 'ocropus-nlbin -o "%s" "%s"/*' % (
os.path.join(job["output_dir"], "tmp", "ocropus-nlbin"), os.path.join(job["output_dir"], "tmp", "ocropus-nlbin"),
os.path.join(job["output_dir"], "tmp", "tiff_files", file) os.path.join(job["output_dir"], "tmp", "tiff_files")
) )
ocropusnlbin_jobs.append(self.addTask(label="ocropusnlbin_job_-_%i" % (ocropusnlbin_job_number), command=cmd, dependencies=split_jobs)) ocropusnlbin_jobs.append(self.addTask(label="ocropusnlbin_job_-_%i" % (ocropusnlbin_job_number), command=cmd, dependencies=split_jobs, nCores=max(1, self.nCores / length(self.jobs)))
### ###
# Task "tesseract_job": perform OCR on binarized images # Task "tesseract_job": perform OCR on binarized images