Optical Character Recognition (OCR) in Python using keras-ocr

In this video, I'll show you how to detect text in images. I will show you how to detect individual words and how to place bounding boxes around them using Python's keras-ocr library.
Link to keras-ocr : github.com/faustomorales/kera...
Link to my notebook : github.com/bhattbhavesh91/ker...
Join this channel to get access to perks:
/ @bhattbhavesh91
If you do have any questions with what we covered in this video then feel free to ask in the comment section below & I'll do my best to answer those.
If you enjoy these tutorials & would like to support them then the easiest way is to simply like the video & give it a thumbs up & also it's a huge help to share these videos with anyone who you think would find them useful.
Please consider clicking the SUBSCRIBE button to be notified for future videos & thank you all for watching.
You can find me on:
Blog - bhattbhavesh91.github.io
Twitter - / _bhaveshbhatt
GitHub - github.com/bhattbhavesh91
Medium - / bhattbhavesh91
About.me - about.me/bhattbhavesh91
Linktree - linktr.ee/bhattbhavesh91
DEV Community - dev.to/bhattbhavesh91
#OCR #Python

[@dannyboi0809]

LYSM BROO... finally keras-ocr worked after trying out numerous other libraries!!!

[@bhattbhavesh91]

Great to hear!

[@prathmeshpatil4456]

Good one!...looks like it’s slightly better than easy-ocr but at the same time a little slower?

[@sivaprakash8148]

Awesome Video Sir...

[@saiprakash5224]

Great video!! Thanks for sharing!!

[@bhattbhavesh91]

Thanks for watching!

[@techspoc7442]

Is it useful in extract table from scanned image PDFs???

[@creativeworld_nm]

Thank you so much for the video. I have a doubt actually , if I wanna store the output in a dictionary with a key value pair, how can I do that as every other document has different columns and values like if I give an invoice pdf converted to image as an input. So how can I deal with this situation?

[@divyadivya3052]

Same situation how to get output like dictionary did u done?

[@brucebuchanan2058]

thanks for the video -- my input images don't work unless I convert them to grayscale first -- how can I convert them to grayscale in my code?

[@JoelPascualMahindo]

Can be used to extract punctuation signs within the text?

[@nadyasudusinghe2213]

Thank you very much

[@ashutoshverma4638]

Can this be used for handwriting scanning.

[@lakshmiharnim6170]

I wanted keras textOCR trained model to extract text from imaged ......
kindly help me with that if possible

[@ankursaggu]

im getting error " 'AxesSubplot' object is not iterable"

[@qzwwzt]

What is this error ModuleNotFoundError: No module named 'tensorflow'

[@govindsheshadri7254]

can you please tell how can use keras ocr in translation example hindi to english urdu etc how can do that if possible make a video on it please

[@yusufsAnt31]

Hello sir. Thanks for video. i got this error: "TypeError: annotate() missing 1 required positional argument: 's'
". How i can fix it? And i have one question more, how i can print these images as text? Thanks

[@jettsetgo9507]

Same problem here

[@amolangre5014]

got the solution?

[@amolangre5014]

@@jettsetgo9507 got the solution

[@krofnoops]

happens when your images array only contains one image. The iteration expects to be able to iterate more than one.

[@ahankamat]

i want to recognise scene text images in multiple languages like spanish , german , chinese , etc. How can i do that?

[@sagihaviv5675]

is there a way we can do sentiment analysis only in ocr, without running nlp

[@bhattbhavesh91]

didn't understand your question!

[@3rnanda]

hello sir. what is the name of editor you are using on this video ?

[@larkinsjeffry1351]

Google colab

[@Zeeshanmys]

Which has better results tessaract or keras..??

[@arnav.w2672]

depends on images. Want to detect faraway images Keras is better. Want to detect close ones, Tesseract it is.

[@RAZZKIRAN]

WILL IT WORK FOR HANDWRITTEN TEXT IMAGES?

[@balatripurasundarivemulako3522]

yes,it is working

[@pasupuletilithin6006]

in which platform u are used

[@bhattbhavesh91]

Google Colab!

[@ludozzlives]

How can I extract the text to a txt or a CSV?

[@truongzgiang]

They only know to show it on colab , you want save to PC you have to use your eyes and fingers for that 😂😂😂 .. true mẹ I'm looking 1000 video on TH-cam , they all only can do the same with this video

[@rajatjaiswal2551]

Create a video on attention ocr

[@bhattbhavesh91]

Thanks for the video recommendation!

[@Redstonedust-rc9nr]

Nic

[@qzwwzt]

1 import keras_ocr
2 import matplotlib.pyplot as plt
File ~\anaconda3\Lib\site-packages\keras_ocr\__init__.py:1
----> 1 from . import (
2 detection,
3 recognition,
4 tools,
5 data_generation,
6 pipeline,
7 evaluation,
8 datasets,
9 config,
10 )
12 __version__ = "0.9.3"
File ~\anaconda3\Lib\site-packages\keras_ocr\detection.py:27
25 import cv2
26 import numpy as np
---> 27 import tensorflow as tf
28 import efficientnet.tfkeras as efficientnet
29 from tensorflow import keras
ModuleNotFoundError: No module named 'tensorflow'