The demand is to locate the invoice within a camera captured image about that invoice. The invoice is always a white paper with printed black or blue characters, tables and red stamps. Sometimes the background behind the invoice is dark(about 60% of all the samples), but others are not. Sometimes there is shadow on it. The question is how to detect the vertices, edges and corners of the invoice in this image? What algorithms should be applied? The android application CamScanner seems to have this function but not is effective every time. So does the CIDetector interface on iOS. What algorithms does CamScanner or iOS use(traditional or deep)?

  • $\begingroup$ Does anyone has any suggestions? $\endgroup$ – piratesailor May 13 at 3:24
  • $\begingroup$ Try datascience.se $\endgroup$ – DuttaA May 13 at 11:29
  • $\begingroup$ @DuttaA Thanks a lot! I have posted it here: datascience.stackexchange.com/questions/51911/… . $\endgroup$ – piratesailor May 14 at 2:08
  • $\begingroup$ @DuttaA Would you please recommend someone to answer my question? $\endgroup$ – piratesailor May 14 at 2:58
  • $\begingroup$ If someone knows the answer they'll answer the question. But I think instead of making the task specific post it as a more general problem like what needs to be done (Probably OCR in this case). It'll increase your chances of getting an answer. $\endgroup$ – DuttaA May 14 at 10:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.