Optical Character Recognition
Deep learning-based vision algorithms read alphanumeric characters on a battery
Manufacturers must be able to locate and decode the alphanumeric codes printed on the bottom sides of batteries quickly and accurately. Specular light and glare can make it difficult for a machine vision system to locate and recognize characters, especially if the characters are deformed. An inspection system needs to tolerate these challenges in order to successfully decode characters.
VisionPro ViDi’s deep learning-based tools locate and read deformed characters, despite image formation challenges. The location tool identifies the region of interest (ROI)—in this case, the top of each cylinder battery, which is marked with an alphanumeric code. The location tool's pre-trained omni-font capabilities recognize characters even if they are obscured by glare and contrast. To train the software, an engineer defines the region of interest on images which contain a representative set of code characters. During training and validation, a technician re-labels only the missed characters until the software’s model correctly identifies all the characters. This deep learning-based approach to OCR saves time during training and development by reducing excessive labeling and ensures accurate reads.