Deep Learning Based Computer Vision Techniques for Defect Detection & Monitoring


The major threat that occurs in pipelines is leakage. The effects of leakage go beyond repair expense and cost of lost oil or gas. It also significantly affects the human lives and environment. In this white paper we have devised a Deep Learning-based solution using Convolution Neural Networks(CNN) and image processing techniques to detect and localize the leakages in oil & water pipelines.

What is in it for you?

The model successfully detects leakages in pipelines even in the presence of limited data. A Deep Learning model in general performs better when the amount of data available is large. Increasing the number of images would result in selecting the optimal hyper parameters for better model stability and reliability.


1 Step 1
FormCraft - WordPress form builder

Beware of fraudulent and fake job offers

It has come to our attention that certain employment agencies and individuals are asking people for money in exchange for a job at ITC Infotech.

Such Agencies/individuals could impersonate ITC Infotech's officers, use the company name/logo, brand names and images illegally, without authorization, and/or try to extract money towards security deposit, documentation processing fees, training fees, and so on.

Please note that ITC Infotech never asks job applicants or members of the public to pay money in any form while recruiting.

Feel free to reach out to us at to report any such incidents that you may have experienced, please use the subject line “Recruitment Fraud Alert” in your message.

Always exercise caution and stay protected against fraud:

  • Do not pay money or transfer funds to anyone toward securing an ITC Infotech job. ITC Infotech will not accept liability for any losses that may have been suffered by the victims of such fraudulent activities.
  • Be careful when sharing your personal information and protect yourself from potential damage. Do not engage with people who fraudulently misrepresent ITC Infotech or its employees/officers and try to solicit payments under the pretext of offering jobs.
View Current Openings
Choose Language »
Don`t copy text!