https://github.com/tentaclepurple/42_leaffliction
👁️ Computer Vision model made with Convolutional Neural Network for leaf disease identification
https://github.com/tentaclepurple/42_leaffliction
computer-vision convolutional-neural-networks deep-learning matplotlib opencv plantcv python streamlit tensorflow
Last synced: 3 months ago
JSON representation
👁️ Computer Vision model made with Convolutional Neural Network for leaf disease identification
- Host: GitHub
- URL: https://github.com/tentaclepurple/42_leaffliction
- Owner: tentaclepurple
- Created: 2024-09-18T19:43:47.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2025-01-10T09:59:43.000Z (over 1 year ago)
- Last Synced: 2025-03-28T18:51:59.707Z (over 1 year ago)
- Topics: computer-vision, convolutional-neural-networks, deep-learning, matplotlib, opencv, plantcv, python, streamlit, tensorflow
- Language: Python
- Homepage:
- Size: 3.21 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## Leveraging AI 🤖 to Detect Plant Diseases 🍃
Now it's time to show the latest development in a plant disease analysis project. By utilizing deep learning techniques, we've created an interactive application that can accurately detect and classify various apple and grape plant diseases.
At the core of our solution is a Convolutional Neural Network (hashtag#CNN) 🧠, a powerful AI architecture that excels at recognizing patterns in images. CNNs work by extracting hierarchical features from the input image, allowing the model to learn and identify complex disease signatures.
The project follows a multi-faceted approach:
### 📊 Data Augmentation:
Recognizing the limited availability of training images, we leveraged data augmentation techniques such as rotation, blurring, flipping, and adjusting brightness and contrast. This helped to expand our dataset and improve the model's robustness.
### 🔍 Image Transformation:
We applied advanced image processing methods, including Gaussian blur, masking, region of interest (ROI) analysis, pseudolandmark detection and histogram. These transformations made possible to extract and analyze key features of the plant leaves, enhancing the disease detection capabilities.
### 🔮 Disease Prediction:
Obtained deep learning models, trained on extensive datasets of plant pathology, can accurately predict the type of disease affecting the plant. The models analyze the leaf's patterns, colors, and textures to make these predictions with an accuracy of 95%.
Also I've developed an interactive application 🖥️ that puts this powerful AI technology in the hands of users. By simply selecting an image, you can explore the effects of different augmentation and transformation techniques, as well as receive a detailed disease analysis. Please feel free to play around with the app (and remember that, in case it's in sleep mode you'll need to wait a little in order to make the app to wake up)
[Go to App (Ctrl + click)
](https://leaffliction.streamlit.app/)
This project is an example of the incredible potential of hashtag#AI in revolutionizing the way we approach plant health and agriculture. 🌱 But the technology behind this project has applications that reach far beyond the realm of agriculture. These powerful AI capabilities can be leveraged across a diverse array of industries and domains.
Coming from a background focused on visual media, I've been able to add a deep knowledge in image processing and manipulation, which has made the development process very enjoyable for me.
Want to dive into the code and technologies themselves? Don't hesitate to dive into the repository or ask me anything.