Transforming the use of industrial drawings

Using deep learning and computer vision techniques, DataSeer can recognize all instances of specific instruments, valves, lines, and other features in an engineering diagram. Through this automation, it reduces the time needed to process any given drawing by up to 90% compared to a typical manual process. It also reduces detection errors to less than 1%, compared to the usual 5-10%.


Applicable Use Cases


Estimating costs for fabrication and construction


Locating symbols and text


Understanding relations in piping and instrumentation diagram


Accelerating verification of QUALITY ASSURANCE & QUALITY CONTROL


Creating line, valve and instruments lists


Isometric data sheet extraction


Extracting information from safety protocol documents


Converting legacy drawings to digital drawings


Information Access
DataSeer’s user interface was created with our users at the forefront. It takes less than one minute to upload up to 100 P&ID’s, and most users are self-sufficient within 15 minutes of their first demo.
Information Access
Using neural network techniques such as object detection and image classification, DataSeer can automatically locate and tag objects across multiple drawings for a user to quickly review.
Information Access
DataSeer’s powerful search functionality allows a user to quickly locate a specific item within their complex set of digitized drawings, removing the need to spend hours manually scanning diagrams.


Information Access
When tested across a pack of 50 drawings, DataSeer has shown a drastic decrease in time spent extracting objects and line numbers by 90%, making a week-long job a matter of a few hours.
Information Access
Human error is unavoidable when manually extracting information from hundreds of complex engineering drawings. With DataSeer, a user can trust the software to extract information while they can manually intervene when needed.
Information Access
With DataSeer, there is no lengthy download process or command-line interaction needed. Simply login via the Google Chrome browser, and be up and running within a matter of minutes.




Industrial MACHINE LEARNING Case Study: Digitization of Engineering Diagrams

Schematic diagrams are the bread-and-butter of the industrial engineer, and some examples include piping & instrumentation diagrams (P&IDs), process flow diagrams (PFDs) and isometric diagrams. Here we review common pain points that the industrial engineer faces when working with these diagrams and explain what you can do to alleviate some of these burdens.

More Info

Case Study

Enhanced Data Extraction From Industrial Drawings

DataSeer, a machine learning application effectively extracting information from P&IDs and other engineering drawings by leveraging deep learning and computer vision techniques Results.

More Info


Do I need to install DataSeer?

No. DataSeer runs in a web browser with no installation necessary.

What web browser should I use to run DataSeer?

You must use Google Chrome, preferably the latest version.

What file formats can DataSeer extract data from?

DataSeer can extract data from PDF files.

What document types can DataSeer extract data from?

DataSeer can extract data from piping and instrumentation diagrams (P&IDs). In the future, we plan to give users the ability to extract data from other types of documents, such as isometric and one-line diagrams.

How many diagrams can I upload per project?

You can upload up to 100 pages of diagrams per project.

Can I search for text on diagrams?

Yes. DataSeer has a powerful text search feature that lets you find text that would be undiscoverable in other applications. To use this feature, simply enter text keywords in the Pattern field during the identification and review process.

Can I export data from DataSeer?

Yes. After you have labeled valves, instruments, and/or line numbers on the diagrams in a project, you can export data for these items into a comma-separated values (CSV) file.

What is the security policy for user data in DataSeer?

DataSeer's security policy is user-centric. DataSeer authenticates each user at login with centralized validation to all information assets against configurable user and system permissions.

DataSeer stores two types of data from diagrams:

  • Structured metadata
  • Document images

Both of which are stored within Arundo’s cloud environment for the life of the project. Stored data and project information can be permanently deleted by the project owner at any time. Access to the stored data is controlled by the project owner within DataSeer, and data access is validated through authentication at each transaction.


GET started with dataseer

Interested in seeing how DataSeer works first hand? Request a demo with one of our experts so you can discuss how to use DataSeer to begin converting your engineering diagrams today.