OCR Engine Developer
Budget: $25 – $50 USD
We have developed a proprietary Mobile Wallet and during the initial signup stage, a customer captures the id image from their phone camera or scan. We want to extract the information from the id and compete the application process automatically for the customer. The customer would review the information to confirm.
Since we are processing government-issued ID from multiple jurisdictions, privacy is very important. We want to have our own OCR engine rather than Amazon Textract or Google Cloud Vision. The OCR engine should come from well-known open resources and libraries such as LSTM, Tesseract, OpenCV, etc.
In our project architecture, the image is uploaded to our server; therefor the OCR engine will be integrated into the back-end rather than the mobile app. The extracted details when then be pushed back with a filled out application form. In this case, we need GPU acceleration to reduce the processing time.
Since we are processing government-issued ID from multiple jurisdictions, privacy is very important. We want to have our own OCR engine rather than Amazon Textract or Google Cloud Vision. The OCR engine should come from well-known open resources and libraries such as LSTM, Tesseract, OpenCV, etc.
In our project architecture, the image is uploaded to our server; therefor the OCR engine will be integrated into the back-end rather than the mobile app. The extracted details when then be pushed back with a filled out application form. In this case, we need GPU acceleration to reduce the processing time.