Plant Disease Detection System Introduction | Image Classification Project | Overview of Project
HTML-код
- Опубликовано: 11 сен 2024
- Hello and Welcome guys
In this project, we'll learn how to make a powerful deep learning model for 38 different classes of image
In this video, we'll see the entire project at a high level
Later in the upcoming video of this playlist, we will tackle and implement the entire project step by step
Playlist Link: • Plant Disease Detectio...
Project Link: github.com/ani...
Dataset Link:
www.kaggle.com...
Sharing you all required libraries with there version
tensorflow==2.15.0
scikit-learn==1.3.0
numpy==1.24.3
matplotlib==3.7.2
seaborn==0.13.0
pandas==2.1.0
streamlit==1.31.1
He'llo sir could u please share the synopsis of this
I love you bro! Thanks for this masterpiece, your other videos are wonderful too. Much love broski
Thanks alot
Keep learning 😊
thank you for this playlist and where can i found the dataset of plants used for agriculture(like groundnut,turmeric,paddy,..etc)
Check on kaggle, uci machine learning repo, mit dataset these are some popular platform where you will get opensource dataset
Sir I am unable to extract the dataset it was structing at 50% please tell how to resolve this or tell me the other dataset similar to this. Please replay sir
Check your extracted folder may be its already there
sir please can you share the list of algorithms other than cnn ..
CNN(Convolution,MaxPooling etc you can get more detail from my CNN tutorial video)
Stochastic Gradient Descent
Adam Algorithm for Optimisation of Neural Networks
Hello, I wanted to know are machine learning and deep learning different? since you have mentioned deep learning in every video of the playlist. Thanks in advance.
Deep learning is subset of machine learning
Where we try to develop learning systems using neural networks
Neural networks is an artificial system
It tries to replicate the some part of functionality of human brain
Where as Machine Learning uses some algorithm to solve problems
There is an error. If we add any other image rather than an image of a leaf the model still predicts and returns an output. Kindly tell me how to resolve this issue. Thank you😊
You can set some threshold like if mac probability score given by model to the class is more than your threshold provided then only give answer else say I can’t say for this etc
Hi please why was open cv not used for image preprocessing or has this been done before, also can VS code be used to build this system
Tensorflow is more powerful as compared to opencv for image feature extraction and it supports lots of modules
Yes you can use vs code
Thank you very much for your reply. When using vs code do you just install the libraries and use the codes from the video tutorials there directly
hi is it cool if i use this for my final year project? and is it possible to get animal disease detection videos too?
is it cool with your professor(s)/instructor(s) ?
Which DL model you have used for this project sir?
Convolutional neural network used to build architecture from scratch
Sir please say which algorithm are used in this program?
Convolutional Neural Network
Thanks bro
Is the playlist fully completed?? I want to make my major project on plant disease detection system
And one more thing can we do the same thing with image segmentation, here you are using image classification but can we use image segmentation will it increases the accuracy and precision?
Yes playlist is completed
Yes you can try just let me know what you are getting from it
Hello. What model did you use? Like is it CNN or?
CNN model from scratch
Hello sir i have a question since i'm consued. So basically we want to only focus on a speciifc plant like (corn) but then we only have a very limited datasets for each of the diseases. What will happen if we experiment or predict a plant that is not part of our programed system,(not part of the given disease in that dataset) woudn't it be useless and how will we able to do that? What are your recommendation please. Thank you sir!
Then you have to collect dataset those are not part of it and re-train your model
Hey, can you please suggest basic prerequisite before starting this?
Prerequisite is knowledge of python
any specific mathematical concepts?
bro i am trying to download the data set that you have provided with is not able to extract the files its stuck on 50 percent what should i do'
Bro, is it still stuck on 50% ?
@@williamsherman6995 no
@@williamsherman6995 No somewhere about 93/94
Just give the directory of the dataset@@williamsherman6995
I think so
sir we are facing difficulties of import the dataset in jupyter it show error of "archive.zip is not UTF-8 encoded" can you solve this problem sir
You have to do extraction first then use respective folder in keras class
@@SPOTLESSTECH i already extract the file sir
Can i use medicinal plant dataset ? To predict medical plant?
Yes
@@SPOTLESSTECHHi, I'm getting very low accuracy for my dataset. Could you help me with that?
Model having low accuracy due to following :
1. Dataset is not good enough for training
2. Make sure each class having good numbers of images there is no bias in the class like one class having high examples and others are comparatively low
3. Check training accuracy and validation accuracy based on that check your model having overfitting or under fitting
Watch my video
Fruits and Vegetables Recognition System Part-9 | Improving Deep Learning Model Performance
ruclips.net/video/FTBVsYg_FVg/видео.html
In this I explained how to avoid the problem of overfitting
Hello sir, I'm working on papaya disease detection but I'm having trouble with data. I have very less data for every class and I'm unable to collect anymore, which is why I'm thinking of augmenting the data that i have already. Could you help me with augmentation or refer any good video or tutorial regarding this? I have watched a lot of augmentation videos but they end up hampering with my data. I'm very short on time, please help.
Try to follow documentation of tensorflow
They also have tutorials along with good blogs
Whats the minimum pictures needed for each class enough for the data
sir can you please help i am facing some error while loading the keras file
Dm me on insta with screenshot
Hello sir, i want to perform this on a specific plant but there is no dataset is present.. i want to create the dataset please suggest how to do it
Collect images of each plant and then map those images with label and you are done then you can proceed like this
Try to put only the plant which you want to predict don’t put unnecessary background
So that model doesn’t learn irrelevant things
For good accuracy
Refer this dataset example and try to crop and put images like this
@@SPOTLESSTECH thank u sir.. but how to map those images with label ?? + Do I need to save those images with disease names ? And do I need to make a separate folder for different diseases
If you make separate folder for each classes that would be great as in keras there are some functions which will take your folder name as label and consider entire thing inside that folder as a data for that label
@@SPOTLESSTECH thank you sir
Sir can I use this project in final year
Please reply 🙏🏻🙏🏻
Yes you can also you can use my Music Genre Classification Project
There you will use CNN audio processing Melspectrogram and other things
That project is superset of this one
@@SPOTLESSTECH SIR can you please help me to make all these projects in Google colab and drive because I didn't have powerful laptop and I will try music genre classification project
Thank for replying sir 🙏🏻🙏🏻
I wanted to use different architecture other than CNN,how can I do it?
Go to keras api and check architecture
What python version you used sir
Python 3.10
Are u on any other social media platform for clarification purposes
Instagram link is there
Will give my Linkedin profile also
www.linkedin.com/in/animesh-py/
is it Deployed on GCP or any other cloud??
Not yet
Bro i need to submit this project as my minor project ..so i need this project documentation...if possible and if you have can you provide that please ..
Bro I don’t have documentation whatever we have discussed in this playlist just go through this and document it
can u provide the documentation or just tell what technologies did you mentioned in ur doc
sir can you help me it give me one prediction "potato healthy" if i pass any picture can you help me
Follow step by step this playlist you will not get error
In your case, it looks like your model is biased
what are the pre requisites for this
Just the knowledge of Python Programming language
Also you have familiarity of how to install libraries and jupyter notebook IDE
which algoritham you used in this ?
Convolutional Neural Networks
@@SPOTLESSTECH thanks
where we get Research paper for this project?
I don’t have any paper just for reference you can check some paper from google
Like this
journalofbigdata.springeropen.com/articles/10.1186/s40537-023-00863-9
Please tell me how this project is different from other plant disease detection systems??
In this project something is unique??
Is it a website or an app.. Can we use the same code in ios also because i dont have a mac.. Can we close the cmd prompt after the comments run successfully
Sorry I am not getting please elaborate
sir can you provide project report also
Sorry I don’t have project Report
Bro i want to develop crop disease detection
Follow playlist and apply concepts there
My kernel keeps dying on accuracy visualization, what could be the problem? I have been trying to create this system.
Please save history and model once its done
After that restart kernel and Continue remaining process by loading history and model
Don’t run training part
It's saying file load error when I try to run the model....it says model is not UTF-8 encoded
sir we have tried this data set is not ideal not classes are unequal better to find other data set
Yes definitely you can make model robust by trying some other dataset or making sampling by modifying this dataset also by adding some more training example, consider equal number of examples in each classes with good quality etc
Can this help me in chromosome identification?
Yes collect dataset first and then apply concepts from here
Bro i want crop disease prediction
Watch this playlist from start to end you will apply these concepts to make model for crop disease prediction
If I get some time then will make project playlist on that also
can u tell me where i can find the the documentation
Follow project playlist from start to end step by step
Here I have explained each and everything from zero level wherever needed i have given the reference of documentation
Can you perform a visualization on the dataset