Vision based recognition

Symbols have been around since recorded history, and they have been used to convey everything from stories of human lives to quantities of goods bartered between kingdoms. Symbols are everywhere – signage to govern traffic, icons to launch apps on our smartphones, and the sequence of alphabets and numbers for troubleshooting technology around us.


The combination of alpha-numerals in a complex arrangement is a cruel joke on memory, not only to remember the right order but also to remember what it signifies (I’m sure we have enough experience remembering passwords and changing them every few months). As we deal with symbols or a cryptic combination of alpha-numerals every day, human errors creeping in are real and the consequences are disastrous.

Under these circumstances, technology does well by mapping symbols to their corresponding meaning, corrective action, or anything else depending on the context and industry. The challenge is translating the symbols without errors, and most sectors find human errors a major cause of concern. People naturally introduce misspells or similar errors if they are forced to read and record thousands of alpha-numerals day-in and day-out. But the same task performed by a machine does not get tired and produces the same result.

Take, for example, the error symbols generated by your car what do you do when you see one of those lights constantly blinking while you drive? You open the dictionary of your car (owner’s manual) that tells you what it means and what you need to do. With a simple point-and-click app, you can migrate that dictionary into your phone.


When you drive your car into the dealership for routine servicing, the service technicians do something similar. They just use a more complicated translation document with far more alphanumeric symbols and multiple outcomes based on the readings. For e.g., if the voltage is 5V, the outcome is different vis-a-vis if the voltage is 2V. Extend this to critical systems like manufacturing/production lines where stoppage in one section of the assembly line stops the entire process, and to be able to capture quality or error while the line is moving saves substantial time, energy, and effort. Or in the energy sector where downtime cuts off the customers on the grid.


Fabrik enables access to industry-leading computer vision algorithms to detect codes and sequences through our platforms. You can swiftly scan any code without any errors. The real-time dashboard captures every piece of information and displays the object in 3D for better visualization and error-proofing . The ability to transform the current management of error codes/symbols into an automated system thereby creating the first steps of quality digital twins is a powerful application for automated management of key assets and processes for the future.

Share Article
Founder of Fabrik and an oblivious jargon-ridden semaphore, you can spot him using flags during zoom calls. His best friend is Google, and he is as funny as the Fermi Paradox.