T-Hub’s United Technologies Innovation Challenge in the Smart Aerospace Sector –
for Machine Vision and Predictive Analytics!

Yes, it is true that Artificial Intelligence helps machines to “see”. This blog simplifies the primary differentiation between Computer Vision (CV) and Machine Vision (MV). Each has strengths, limitations, and best use case scenarios for these overlapping technologies.

Computer Vision: A Brief History

When computer vision started to take shape in the 1960s, its aimed to mimic human vision systems and asked computers to describe what it saw, automating the analysis process. This technology is the precursor to artificially intelligent image recognition. Prior to this, image analysis had to be done manually — , from x-rays to and MRIs to hi-res space photography.

Both computer vision and machine vision use image capture and analysis to perform tasks with speed and accuracy that is unmatched by human eyes. Just like animals, computers “see” the world differently from humans — counting the number of pixels, discerning the borders between objects by measuring shades of colour, and estimate estimating spatial relations between objects. With this in mind, it’s more productive to describe these related technologies using commonalities — distinguishing technology with specific use cases rather than differences.

Computer vision and machine vision systems share most of the same components and requirements:

  • An imaging device containing an image sensor and a lens.
  • An image capture board or frame grabber may be used (in some digital cameras that use a modern interface, a frame grabber is not required)
  • Application-appropriate lighting
  • Software that processes the images via a computer or an internal system, as in many “smart” cameras

So, what’s the actual difference between computer vision and machine vision?

Computer vision: It refers to automation of the capture and processing of images, with thorough image analysis. In other words, computer vision does not just see, but processes and provides output through image recognition.

Machine vision: It refers to the use of computer vision in industrial & especially and manufacturing environments &. It is a subcategory of computer vision.

Computer vision in action

As computer vision evolved, algorithms were programmed to solve individual challenges, and these become got better at doing a job on repeating a task. Correspondingly, we saw a rise in improved deep learning techniques and technology. With deep learning, we’re now able to program a supercomputer to train itself, self-improve over time and provide portions of these capabilities to businesses as online applications, like cloud-based apps.

For these machines to learn, they need to be fed data.

Machine vision and the smart aerospace

The ability to visually identify issues like product defects and process inefficiencies is critical for aerospace manufacturers to optimise cost efficiency and drive high customer satisfaction. Since the ’90s, machine vision systems have been installed in thousands of factories worldwide, where these are used to automate essential quality assurance functions and improve efficiency. With enhanced data sharing capabilities and improved accuracy powered by innovative cloud technologies, the use of machine vision driven systems in manufacturing has accelerated. Manufacturers realize that machine vision systems are essential investments for meeting goals around quality, cost, and speed.

Machine vision on the production line in the manufacturing environment

Defect detection and resolution are integral to any manufacturing process. Companies are implementing machine vision solutions to address the root cause of defects. By installing cameras on production lines and training machine learning models to identify complex variables that define a good or bad product, it’s possible to identify defects in real time and determine where these occur in the manufacturing process, so that proactive steps can be taken.

Annotating a machine learning model for vision technologies

To achieve your computer or machine vision goals, you first need to train the machine learning models that make your vision system “intelligent.” And for your machine learning models to be accurate, you need high volumes of annotated data, specific to the solution you’re building. There are free, public-use data sets available that work well for testing algorithms or performing simple tasks, but for most real-world projects to succeed, specialised data sets are required to ensure these contain the right metadata.

Smart Inventory

For manufacturing distribution, though ERP systems have helped in establishing accurate forecasting to implement analysis, still does not provide inventory managers with a complete view on the optimisation.

But when analytics solutions like predictive analytics immerse with an ERP system, it acts as game changer in the inventory optimisation process & aids industry in the following areas:

  • Maintaining the minimum cost of average on-hand inventory (no or very little end of month reduction for financial reporting),
  • Achieving highest desired fill rates for the items being optimised
  • Using least cost reorder frequency
  • Establishing predictive control over inventory performance using a company’s IT/ERP system

T-Hub’s United Technologies Innovation Challenge

T-Hub, which leads India’s pioneering innovation ecosystem that powers next-generation products and new business models, and United Technologies Corp. (UTC), a leader in aerospace and building technologies, launched the UTC Innovation Challenge.

The challenge targets startups working on machine vision and predictive analytics for the aerospace industry. Shortlisted, scalable startups will receive support from technical subject matter experts and business mentors from T-Hub and UTC.

Additional details, including how to apply, are outlined here

If you’re a startup working in machine vision and and predictive analytics; And would like to partner with UTC, a Fortune 500 company headquartered in New York, helping UTC in its use cases through a PoC, then don’t miss this opportunity. Startups operating across the globe are invited to participate. Apply today.

In case of any queries reach out to Aditya Patro, Program Manager, T-Hub, Corporate Innovation at aditya.patro@t-hub.co / +919966775875

Note: This article written by Aditya Ranjan Patro, is reproduced on this platform from here.

Enabling and empowering an ecosystem hungry for innovation