Here is the result from the reddit
the numbers above were correctly recognized with a confidence score of 93.74%
Our software can batch infer images which are requiring OCR.
Model infer speed:
Our OCR model achieves an inference speed of 300 pcs/s on an NVIDIA RTX 4060 GPU, and batch processing speed of 1,250 pcs/s (image size: 640x640 pixels)
Software download link: https://www.dlcv.com.cn/
Our software offers free annotation, training, and testing features
Due to the OCR model is not free and the software interface is currently available in Chinese only , you can also email me some images, and I’ll process them through our OCR model and share the results with you, it’s free! Please feel free to reach out to me directly or leave a comment below this post
My email : luqi@dlcv.ai