The AI-based OCR system improves quality control and traceability

     Cognex has expanded the capabilities of its In-Sight 2800 vision system to incorporate OCR. The new ViDi EL Read tool is powered by edge learning technology that easily interprets human-readable characters on reflected, low-contrast, and non-flat surfaces. "In-Sight 2800 with edge learning redefines ease of use, making AI-based applications accessible to anyone regardless of experience level," said Lavanya Manohar, vice president of Vision Systems at Cognex. "

     ViDi EL Read replaces complex programming with exam-based training, making it simple to model OCR and simultaneously read characters on challenging surfaces or even multiple lines of text." Easy-to-use OCR solutions help food and beverage manufacturers easily read expiration dates, even on curved surfaces, to verify freshness and deter consumer recalls.

     Medical device and pharmaceutical companies can verify dates and lot numbers to provide vaccine efficacy and compliance with medical standards. Electric vehicle manufacturers can quickly locate and read alphanumeric text etched by lasers on the bottom of electric vehicle batteries to improve traceability, and logistics facilities can decipher codes and text on various package types to provide proper routing and stop rework.

     Setting up OCR applications is time-consuming and requires hours of programming by trained engineers, which prevents many companies from automating this type of inspection. Using the tool, models can be set up and deployed on the device in a matter of minutes with just 10 sample images. The accuracy of the results is measured by the confidence score displayed below each training character, which provides real-time visual feedback and ensures reliable traceability throughout the supply chain.