AI Powered Visual Inspection Platform Launched
Qualitas Technologies has launched one of the first AI-powered Visual Inspection Platforms to provide a powerful deep learning operations (DLOps) workflow for Industrial Vision Inspection.
Visual Inspection products using AI have been making headlines with technologies like deep learning gaining popularity due to the ease of programming and maturity in the platforms used to develop solutions. However, developing production-ready solutions is still very complex and time-consuming. Much effort goes into building ‘Pipelines’ or DL Ops workflows. Qualitas Technologies has been providing complex solutions as a system integrator for over 10 years when the team realized that the workflow for developing AI-based solutions had a lot of common tasks and steps which could be greatly simplified through technology leading to the development of the Qualitas EagleEye platform.
System Integrators or project managers in manufacturing companies can leverage the EagleEye platform for tasks like data collection, ground truth annotation, image annotation, and deep learning solution development. Furthermore, the platform enables training and evaluating powerful and accurate state-of-the-art deep learning solutions, at one tenth of the time it would take to develop something with open-source tools and platforms.
Development of machine vision solutions using deep learning takes weeks of development in a lab environment before deployment to production. After which, it could sometimes take up to a year to ‘fine-tune’ models to the desired levels of accuracy. The QualitasEagleEye, with its optimized workflow-driven application, can help users develop production-grade solutions in a matter of weeks with the help of just a browser.
Employing a 4-step process, solutions are designed and developed without the investment of specialized AI training hardware and software. Solutions common to visual inspection in manufacturing like surface defect identification, optical character reading, part, and assembly verification, and robotic guidance can be developed and deployed in production lines in a matter of weeks, consuming limited training images.
Speaking about the development, Raghava Kashyapa, CEO and Founder of Qualitas Technologies said, “Reducing the complexity and cost barrier to train and deploy deep learning-based machine vision systems has been a long-standing need for the industry. System integrators and solution developers often take months if not years to fine-tune production-grade solutions and we’re happy to have achieved 10x simplicity with the launch of the QualitasEagleEye Visual Inspection Platform. Integrators and machine builders can take advantage of the fully managed and LEAN integration workflow to develop solutions which were prohibitively expensive and time-consuming owing to complexity in software and hardware infrastructure required for both training and inference.”
Machine Vision systems have been proven to be a direct replacement for human decision making and with the advent of Deep Learning, these systems are able to achieve better than human inspection accuracy.
For more information: www.qualitastech.com
