Mud Log Image Extraction Tool Bug Fixes and Code Clean Up

Extraction tool can batch process mud log images, save results to the database, and even export images with the marks drawn on the images.

In this challenge we need to update the tool with two changes:

1. We need to fix the High and Medium issues from the PDF, as indicated in this screenshot: http://take.ms/t8yEo
2. We have some mud logs in which the logging sections of the documents are interrupted by summary sections. When we run the remove-outlier command we're removing the phrases from the headers and footers of the documents.  But we're not removing tags/phrases that are part of the non-logging sections which are in the middle of the documents. We need to change this to remove these phrases from the middle of the documents and improve the accuracy of our counts. Here is a sample of one of these sections:
   
   As you can see we've flagged 2 shows above but these shows aren't really in the logging sections of the document.
   
   We're receiving a LIC file for each of sections, so I think what we need to do when were processing the LIC files is track these sections and do a more sophisticated job of tracking the active logging sections of the document. If a phrase is not in the range defined in any of the lic files, the phrase should be removed.

You need to make sure your changes don't break any of the existing features. We'll provide some test images for your development but for verification we'll use more images to verify your solution.

Final Submission Guidelines

- Updated solution with the above-mentioned requirements covered
- Detailed readme on how to configure, run and verify your solution.
- Winner may be asked to send a merge request to our repo