aditya9211 / blur-and-clear-classification Goto Github PK
View Code? Open in Web Editor NEWClassifying the Blur and Clear Images
Classifying the Blur and Clear Images
Hello? I'm working on blur image classification using the network on this site.
I have 2,500 images, and I have blur image and clean image for classification, but when I train,
the validation dataset accuracy is not more than 70% and I have a stagnant problem
Please tell us how can improve our validation accuracy. and I want to know the hyper parameter setting
for your neural network you set.
Thank you
Hello, someone else realized a version of this project to run in Colab? I am really interested to try it!!
Dear
I am working on distinguishing blurred images problem. And I think this repertory is interesting and it must be helpful for my work. But I have not yet seen the database used. Could you please tell me which database the repertory had used?
Appreciate your help.
Jasonsey
want to add predict.py for 1 image to predict blur or clear
and addition to test on the user-defined dataset, not from trained splitter one.
Hi - this is great! I've trained my model with 100 blurry and 100 clear and everything seems to be doing well - 96% accuracy, but now for the prediction I can only get it to run one image at a time to give me the good or bad label - am I missing something? I have about 7,000 images to label and I'm looking to set it up for an influx of daily imagery.
The IMAGE_PATH doesn't accept *.jpg or directing it to a folder of images... can you advise?
And then is there an easy way to move the predicted image to a folder based on its label? Trying to get the non-blurry photos to be sent into another NN to be put into a classification model.
Thanks!
Hi, where are your blur images samples from? Is blur images from some algorithm generated or download a collected image dataset or by google crawler?
Thanks.
Current the orignal size is (320,240) but it needs to be fed from the user though.
command
pip -r install requirements.txt
failed when using python > 3.6+
Hi thanks for sharing your work.
but i get this error when running train.py
Train Data Path Success .....
Train Label Path Success .....
Test Data Path Success .....
Test Label Path Success .....
Saving the splitting results......
Traceback (most recent call last):
File "train.py", line 564, in <module>
main()
File "train.py", line 532, in main
train_images = np.insert(train_images, 0, 1, axis=1)
File "/home/jeniffer/.local/lib/python2.7/site-packages/numpy/lib/function_base.py", line 4431, in insert
new = empty(newshape, arr.dtype, arrorder)
MemoryError`
Any ideas?
combined_img = nd.median_filter(combined_img,radius)
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.