YukSing12 / Steel-Billet-Character-Recognition

3 stars 0 forks source link

English | 简体中文

Steel Billet Character Recognition,SBCR

Description

Steel Billet Character Recognition

Overall directory structure

The overall directory structure of SBCR(Steel Billet Character Recognition) is introduced as follows:

Steel-Billet-Character-Recognition
├── Xiang-Steel-Billet-Dataset (Not open source yet)
├── Tangshan-Steel-Billet-Dataset (Not open source yet)
├── PaddleOCR
├── PaddleSlim
├── TCPServing
└── output

Todo list

Installation

Requirements:

Installation:

Create conda environment

  conda create -n SBCR python=3.7.4
  conda activate SBCR

Install PaddlePaddle with CUDA 10.0 (more version):

  conda install paddlepaddle-gpu==2.0.0rc0 cudatoolkit=10.0 -c paddle

Check installation of paddle

  CUDA_VISIBLE_DEVICES=0 python ./run_check.py

If installed successfully, it will show that

  Your Paddle Fluid works well on SINGLE GPU or CPU.
  Your Paddle Fluid works well on MUTIPLE GPU or CPU.
  Your Paddle Fluid is installed successfully! Let's start deep Learning with Paddle Fluid now

Clone Repo

  git clone https://github.com/YukSing12/Steel-Billet-Character-Recognition.git
  cd Steel-Billet-Character-Recognition/

Install PaddleOCR-develop from repo

  git clone https://github.com/PaddlePaddle/PaddleOCR.git

Install paddleocr

  cd PaddleOCR/
  pip install -r requirements.txt -i https://mirror.baidu.com/pypi/simple
  python setup.py install

Clone PaddleSlim-develop from repo

  git clone https://github.com/PaddlePaddle/PaddleSlim.git

Install paddleslim

  cd PaddleSlim/
  pip install -r requirements.txt -i https://mirror.baidu.com/pypi/simple
  python setup.py install

Dataset

Xiang-Steel-Billet-Dataset

Tangshan-Steel-Billet-Dataset