Text Detection using Neural Networks | OPENCV Python
[ad_1]
In this video we are going to learn how detect text by creating a Convolution Neural Network to Classify digits from 0 to 9. We will write the Training code from scratch and go step by step to train about 10000 images of 10 different classes . Later we will create a testing script to use along with a webcam. I will be sharing all the details including the version of the packages and the parameters for the training
Code & Text Based Version:
(Code and be found in the Advance Topics section)
https://www.murtazahassan.com/courses/opencv-projects/
Follow Me:
Instagram : https://www.instagram.com/murtazasworkshop/
Facebook: https://www.facebook.com/murtazasworkshop
Website: https://www.murtazahassan.com/
Github: https://github.com/murtazahassan
Thingiverse: https://www.thingiverse.com/Murtaza_Hassan/
################################################
Full OpenCV 3 Hour Course: https://youtu.be/WQeoO7MI0Bs
################################################
#CNN
#Keras
#DigitRecognition
Like — Comment — Share
Useful Links:
DC Motors + Wheels + Chassis: https://amzn.to/2SCZon3
DC Motors + Wheels: https://amzn.to/2QeEusw
Arduino UNO: https://amzn.to/2QdKthB
Motor Driver: https://amzn.to/35grl6x
Battery: https://amzn.to/2Fadc0c
Recommend Webcam for Computer Vision: https://amzn.to/2MNtVKZ
Budget Webcam: https://amzn.to/2ZP47Ug
Cheap Drone for OpenCV: https://amzn.to/2TZpsJy
Raspberry Pi 4 Best Starter Kit: https://amzn.to/2YVCEkL
Raspberry Pi Recommended Battery: https://amzn.to/2C0I9pl
Raspberry Pi Camera: https://amzn.to/2BDFcLF
Raspberry Pi Wide Lens: https://amzn.to/2NQH5qD
More powerful Raspberry Pi alternative: https://amzn.to/3eY3AWN
Links:
Classify Traffic Signs: https://youtu.be/SWaYRyi0TTs
How To Install OPENCV in Python:
https://youtu.be/CJXIjApHYVs
Computer Vision Course:
https://www.youtube.com/watch?v=CJXIj…
5 Must Know OpencCV Functions:
https://youtu.be/7kHhz7nkpBw
Easy Object Following Robot using Arduino and PixyCam:
https://youtu.be/w_krOCBk1DE
Robot Arm Arduino Tutorial | Gesture Controlled:
https://youtu.be/gmz7eOB-tCg
Facing any Errors or want to discuss more?
✔️ Join my Discord
https://discord.gg/bFVMkzf
Source
[ad_2]
hi, Murtaza. I have a problem to extract data from plot images those old paper with grid lines and scanned . I know all manually programs.but i have many image. may ı extract data automatically??? Thanks in advance ..
cant install tensor flow no matter how many time i tried to close and open the settings
Excellent effort! I am getting following error at 48:11 of video: NameError: name 'sequential' is not defined. Is it something to do with Tensorflow?
I'm getting this error:
pickle.dump(model,pickle_out)
TypeError: can't pickle _thread.RLock objects
does anyone know why?
I don't know why you are having very less views likes and comments. I just want you to know that you're are doing great work, Keep doing.
Thank you
Great chanel! Thank you!
Question: what minimal hardware would you recommend for running all this OpenCV/TensorFlow tasks?
Somaone has a problem where the graph "loss" and "accuracy" doesnt show anythin?
Salam aleykum! Can anybody explain, what is wrong with np.where(y_train == 0) ? I had future warning.
how can i get this source code sir?
Hi Murtaza, thanks for the tutorial. Is it possible to extend this to include handwritten texts? Do we need additional package?
I am getting below error while pickle the data "typeerror: cannot pickle _thread.Rlock " and graph for Loss & accuracy is generated, but value is not plotted (no lines)
please help Traceback (most recent call last):
File "C:/JustDoIt/Python/mydata/number recognition.py", line 9, in <module>
myList = os.listdir(path)
FileNotFoundError: [WinError 3] The system cannot find the path specified: 'JustDoIt'
Can you show me the source of all your pictures of number? Or you take them by your camera? Thank you so much
One of the reason I like this channel is explanation in clear ,consider manner ,not so fast perfect for beginners like me
hello, how do i get the "my data" folder to show i cant seem to find it
sir how to get code u wrote in above video ??? in link given in description, code is not opening plz help
where can i get these datasets for testing purpose?
can't we use model.save() to save our model in this program?
Thanks Murtaza for this great video. I'm a big fan and I've followed a lot of your projects on your website. My question is how can we extend this to double digits numbers like 10, 11 …. . Do you have an available dataset for that?
Hello sir, mymodel doest run until complete the epochvalue which is set as 10 but its only run 131/2000 stepsperepochval, showing me this error:
WARNING:tensorflow:From E:/PycharmProjects/Number Classfication/main.py:144: Model.fit_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.
Instructions for updating:
Please use Model.fit, which supports generators.
Hello, does it predicts only black digits? Can it work with digits that are with different color?
Hello, I really liked your course. I stumble on error however, can you help me with it?
Traceback (most recent call last):
File "OCR_CNN_Training.py", line 190, in <module>
pickle.dump(model, pickle_out)
TypeError: cannot pickle '_thread.RLock' object
Thank you… However could not locate image folder… I tried git also. I am missing something please help….. Thanks in advance.
Will it work on Mac os
Simplified explaination but Where i can get the umage folders
Nice tutorial but the test part is giving error at cv2.resize line in while loop
from where can i get the data set of images?
Can you provide a link for those images please
Can i get this dataset Please 😍
Getting this error: Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches
Is there any way for loading model in c++ opencv?
Thanks.