Due to an increase in the number of customer inquiries, our front-end processing of original pages with OCR to extract text information has also significantly increased in frequency.
We are now using AI-powered tools and neural network inference frameworks to significantly improve the clarity of images, greatly improving the accuracy of recognition and accelerating our quotation speed for standard engine models.
At the same time, we are also able to correct and process information from nameplate photos with poor image quality — allowing us to quickly and accurately recognise part details without spending extra time going back and forth with customers.
We hope that each step in our technical improvement will bring our customers a better service experience.