OCR (Optical Character Recognition, optical character recognition) is a commonly used automatic recognition technology, widely used in text recognition, automatic processing and computer vision and other fields. In this post, we will introduce how to use the OpenVINO library to easily implement PaddleOCR real-time inference.
The first and second phases of the OCR series have introduced how to use PaddleOCR for text recognition. In these two issues, we mainly introduced how to use PaddleOCR for static image recognition. However, in fact, OCR application scenarios are very rich, including the recognition of dynamic images. Therefore, in this article, we will introduce how to use the OpenVINO library to implement PaddleOCR's real-time inference on dynamic images.
1. First, we need to install the OpenVINO library. It can be installed in Ubuntu or Debian operating systems with the following command:
sudo apt-get install python-dev python3-dev build-essential openvino-io-1.0-cp37-none-linux-gnueabihf
2. Next, we need to import the OpenVINO library and related Python libraries. The following code can be used:
import paddle
import paddle.vision.transforms as T
import paddle.vision.models as models
import paddle.vision.datasets as datasets
import paddle.vision.optimizer as optim
This article is published by mdnice multi-platform