We will be using a python wrapper for Tesseract OCR called tesserocr. You can also look at this tutorial for more information.
- Create an environment with default packages installed.
$ conda create -n tesserocr_env python=3.7
$ conda activate tesserocr_env- Install tesserocr package
$ conda install -c conda-forge tesserocr- Install ipykernel package for adding your conda environment to your jupyter notebook
$ conda install -c anaconda ipykernel-
In terminal, navigate to the folder that contains the environment.yml file.
-
Create an environment with all the required packages installed.
$ conda env create -n tesserocr_env -f environment.ymlFor more details, please refer to managing conda environment.
You may refer to add conda environment to jupyter notebook and remove virtual environment from jupyter notebook.
$ python -m ipykernel install --user --name=tesserocr_env- Go to your home directory.
$ cd ~- Start a jupyter notebook server.
$ jupyter notebook- Navigate to your project directory.
- Create a jupyter notebook under tesserocr_env.
For those who cannot use conda, pip or apt-get to install packages, they need to download the whole environment folder from github, and then append the "path to modules" to sys.path.
sys.path explanation:
When you start a Python interpreter, one of the things it creates automatically is a list that contains all of directories it will use to search for modules when importing. This list is available in a variable named sys.path. That is, sys.path tells the Python interpreter where to import modules.
- Download the whole environment folder tesserocr_env from github and put it somewhere inside the project folder.
- In your script that runs first, include the following codes at the beginning.
import sys
if "../tesserocr_env/lib/python3.7/site-packages" not in sys.path:
sys.path.append("../tesserocr_env/lib/python3.7/site-packages")Directory tree structure of this project:
'''
/tesserocr
├── src
│ ├── example.ipynb
│
└── tesserocr_env
├── lib
├── python3.7
├── site-packages
'''Therefore, when navigating from example.ipynb, "../tesserocr_env" is the environment folder. Note that we need to navigate exactly to the site-packages folder. Hence, the path is "../tesserocr_env/lib/python3.7/site-packages".
- Create a PyTessBaseAPI object and assign it to a variable.
with PyTessBaseAPI(path='/Users/michael/opt/anaconda3/envs/tesserocr_env/share/tessdata/', lang='eng') as api:
# Usage below...- Tesseract API methods are called via this class object.
# Usage below...
print(api.GetAvailableLanguages())
for img in images:
api.SetImageFile(img)
print(api.GetUTF8Text())
# Tesseract API methods are called in this way.
# api.Xxxxxx
# api.Yyyyyy
# api.ZzzzzzPlease refer to my jupyter notebook example.
Image inputs may be rotated. A series of image inputs may have different orientations. For example, "input/pdf2image3.jpg" and "input/pdf2image4.jpg" have different orientations.
images = ["input/pdf2image3.jpg", "input/pdf2image4.jpg"]By specifying psm=PSM.AUTO_OSD, the orientation of all images is automatically detected. Image-to-text then works perfectly under correct orientation.
with PyTessBaseAPI(psm=PSM.AUTO_OSD) as api:
for image in images:
api.SetImageFile(image)
# There are other page segmentation modes (PSMs):
'''
0 : OSD_ONLY: Orientation and script detection only.
1 : AUTO_OSD: Automatic page segmentation with orientation and script detection. (OSD)
2 : AUTO_ONLY: Automatic page segmentation, but no OSD, or OCR.
3 : AUTO: Fully automatic page segmentation, but no OSD. (default mode for tesserocr)
4 : SINGLE_COLUMN: Assume a single column of text of variable sizes.
5 : SINGLE_BLOCK_VERT_TEXT: Assume a single uniform block of vertically aligned text.
6 : SINGLE_BLOCK: Assume a single uniform block of text.
7 : SINGLE_LINE: Treat the image as a single text line.
8 : SINGLE_WORD: Treat the image as a single word.
9 : CIRCLE_WORD: Treat the image as a single word in a circle.
10 : SINGLE_CHAR: Treat the image as a single character.
11 : SPARSE_TEXT: Find as much text as possible in no particular order.
12 : SPARSE_TEXT_OSD: Sparse text with orientation and script detection
13 : RAW_LINE: Treat the image as a single text line, bypassing hacks that are Tesseract-specific.
'''All interfaces can be found in the Python wrapper around the Tesseract-OCR C++ API.
This module provides a wrapper class :class:PyTessBaseAPI to call Tesseract API methods.
This module also provides other helper functions.
def InitFull(self, path=_DEFAULT_PATH, lang=_DEFAULT_LANG,
OcrEngineMode oem=OEM_DEFAULT, list configs=[],
dict variables={}, bool set_only_non_debug_params=False):
"""Initialize the API with the given parameters (advanced).
It is entirely safe (and eventually will be efficient too) to call
:meth:`Init` multiple times on the same instance to change language, or just
to reset the classifier.
Page Segmentation Mode is set to :attr:`PSM.AUTO` after initialization by default.
Args:
path (str): The name of the parent directory of tessdata.
Must end in /.
lang (str): An ISO 639-3 language string. Defaults to 'eng'.
The language may be a string of the form [~]<lang>[+[~]<lang>]* indicating
that multiple languages are to be loaded. Eg hin+eng will load Hindi and
English. Languages may specify internally that they want to be loaded
with one or more other languages, so the ~ sign is available to override
that. Eg if hin were set to load eng by default, then hin+~eng would force
loading only hin. The number of loaded languages is limited only by
memory, with the caveat that loading additional languages will impact
both speed and accuracy, as there is more work to do to decide on the
applicable language, and there is more chance of hallucinating incorrect
words.
oem (int): OCR engine mode. Defaults to :attr:`OEM.DEFAULT`.
See :class:`OEM` for all available options.
configs (list): List of config files to load variables from.
variables (dict): Extra variables to be set.
set_only_non_debug_params (bool): If ``True``, only params that do not contain
"debug" in the name will be set.
Raises:
:exc:`RuntimeError`: If API initialization fails.
"""
cdef:
bytes py_path = _b(path)
bytes py_lang = _b(lang)
cchar_t *cpath = py_path
cchar_t *clang = py_lang
int configs_size = len(configs)
char **configs_ = <char **>malloc(configs_size * sizeof(char *))
GenericVector[STRING] vars_vec
GenericVector[STRING] vars_vals
cchar_t *val
STRING sval
for i, c in enumerate(configs):
c = _b(c)
configs_[i] = c
for k, v in variables.items():
k = _b(k)
val = k
sval = val
vars_vec.push_back(sval)
v = _b(v)
val = v
sval = val
vars_vals.push_back(sval)
with nogil:
try:
self._init_api(cpath, clang, oem, configs_, configs_size, &vars_vec, &vars_vals,
set_only_non_debug_params, PSM_AUTO)
finally:
free(configs_)
def Init(self, path=_DEFAULT_PATH, lang=_DEFAULT_LANG,
OcrEngineMode oem=OEM_DEFAULT):
"""Initialize the API with the given data path, language and OCR engine mode.
See :meth:`InitFull` for more initialization info and options.
Args:
path (str): The name of the parent directory of tessdata.
Must end in /. Uses default installation path if not specified.
lang (str): An ISO 639-3 language string. Defaults to 'eng'.
See :meth:`InitFull` for full description of this parameter.
oem (int): OCR engine mode. Defaults to :attr:`OEM.DEFAULT`.
See :class:`OEM` for all available options.
Raises:
:exc:`RuntimeError`: If API initialization fails.
"""
cdef:
bytes py_path = _b(path)
bytes py_lang = _b(lang)
cchar_t *cpath = py_path
cchar_t *clang = py_lang
with nogil:
self._init_api(cpath, clang, oem, NULL, 0, NULL, NULL, False, PSM_AUTO)- Remember to activate the conda environment first (tesserocr_env).
$ conda activate tesserocr_env- Install poppler.
$ conda install -c conda-forge poppler- Install pdf2image.
$ pip install pdf2image- Take note of the path to a file called "pdfinfo". In my computer, it is stored in "/Users/michael/opt/anaconda3/envs/tesserocr_env/bin".
In this project, "pdfinfo" is stored in here.
Later in the "convert_from_path" function, we have to specify the "poppler_path" argument.
For example,
images = convert_from_path('input/pdf2image.pdf', poppler_path="/Users/michael/opt/anaconda3/envs/tesserocr_env/bin")In the following example, we convert a 3-pages pdf in /tesserocr/src/input to 3 images and save them to /tesserocr/src/input.
from pdf2image import convert_from_path, convert_from_bytes
from PIL import Image
from pdf2image.exceptions import (
PDFInfoNotInstalledError,
PDFPageCountError,
PDFSyntaxError
)
'''
We need to specify the [poppler_path] variable to locate the [pdfinfo] executable (from poppler installation).
That is, [poppler_path] is the path to the folder containing [pdfinfo].
Directory tree structure of this project:
/tesserocr
├── src
│ ├── example.ipynb
│
└── tesserocr_env
├── bin
├── pdfinfo
Therefore, when navigating from example.ipynb, "../tesserocr_env/bin" is the folder containing pdfinfo.
'''
## input pdf file from /tesserocr/src/input
images = convert_from_path('input/pdf2image.pdf', poppler_path="../tesserocr_env/bin")
for i in range(len(images)):
image = images[i]
# image.show()
## save files in /tesserocr/src/input
image.save("input/pdf2image"+str(i+1)+".jpg")