Back to cases

Parcel Processing System

Request a Quote
Categories: Machine Learning INDUSTRY: Logistics
Download file

Problem

The bar codes often get mechanical scratches or any other damages before reaching the sorting point. Defects obstruct the correct identifying of a parcel. The client needed to prevent the possible mistakes made by damaged bar codes and the human factor.

Solution

CHI Software’s team trained the custom neural network to localize the bar code even if any mechanical damages appear. 

Enriched by the software, the automated parcel delivery processing system allows performing a lot of operation, such as:

  • Proper postal label detection among several others;
  • Extraction of country and zip codes and text, if the bar code is not readable;
  • Compositing of the delivery information;
  • Checking the postage — the information is immediately transferred into the database; 
  • Creating the instructions for automatic sending of the parcel to the proper vehicle for delivery.

Result

After the solution implementation, our client has got easy-to-use, powerful automation of the business process. Thru overall interaction, the work of distribution centers worldwide was optimized. Our solution helped the client reduce costs and ensure that parcels, letters, and other items are identified automatically and sorted properly.

Technologies

Optical Character Recognition with Tesseract, Python, С++, NumPy, Pillow, OpenCVscikit-image, TensorFlowKerasCNN, RNN, Darknet, Yolo, CNN-RNN-CTC​.

650

projects

500+

experts

304

clients

4

offices

AI

R&D center

Let’s Talk!

We are glad to see you here. Let's get in touch!

    img img Successfully applied!