people_identifier

Problem Statement:

• Nowadays in many offices, when new candidates gets employed, their faces are checked several times.

• First they check if the faces in hall ticket is same as in college identity card. If the faces match, they continue, else they are kept aside.

• Then, during interview, screenshots are being taken, to check if that candidate's face matches with the hall ticket.

• Once their faces matches everything, they get verified.

• However, this job is done by a staff in an office and imagine the staff verifying 100 of candidates.:face_with_head_bandage:

• To avoid so much stress, I made a small python code that verifies if the candidates faces matches:

  1. hall ticket and college identity
  2. hall ticket and screenshots in interview

Requirements:

• miniconda (get it from here)

Steps:

1. Install miniconda

   

2. While installing it, tick the option that says Add Miniconda3 to my PATH environment variable

3. After installation, go to the place with directories of different people

   

4. Place all the codes there

   

5. Make sure that the details of each candidate is in the directory details

   

6. Inside directory of each candidate, make sure these PDF files exist:

   • college_id.pdf (college id of the candidate)
   • hall_ticket.pdf (hall ticket of the candidate)
   • interview.pdf (screenshots of the interview)

7. Now once everything is verified, go back to the place where the codes exist

8. Type cmd from there

9. Hit enter and the command prompt will open

   

10. Type conda activate base and hit enter. It will activate miniconda.

    

11. Type conda create --name pi --file requirements.txt as above

    

12. Type y as above

    

13. If it ends like this, then we can continue. If it ended up with an error then you have to start from first :(

14. After everything is done, type conda activate pi like below

    

15. Once activated, type pip install opencv-python as below

    

16. After installation, we are good to run the code :)

    

17. Type python candidate_verifier.py and hit enter. Wait for a minute. It will end up like the above if no error occurred.

    

18. Click the candidate.txt to see the output.

    

19. You can import it as a CSV in excel for a better view :)

Troubleshooting:

• Make sure you followed the steps correctly

• Make sure you installed opencv correctly (see step 15)

• Make sure that the 3 files is there in each candidate directory

• If one of the files also doesn't exist, it will throw up an error like the below: