Video redaction usecase

Hello friend, I am newbie to both Ubuntu and Nvidia tech. I have a use case where I want to redact video with financial information on it especially “Numbers”. Based on some AI search, my finding is to leverage the following tech.
DeepStream SDK for video processing
GStreamer for pipeline management
TensorRT for optimized inference
CUDA for GPU acceleration
nvOCDR plugin for text detection

Do you think that is the right approach? Is there someone who has done similar redaction work? Any guidance is much appreciated. thank you.

Jagan.