Update README.md
Browse files
README.md
CHANGED
|
@@ -1,50 +1,92 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
-
# Flower Species Classifier
|
| 5 |
|
| 6 |
-
|
| 7 |
-
Flower Species CNN Classifier
|
| 8 |
|
| 9 |
-
## Model
|
| 10 |
-
Convolutional Neural Network (CNN)
|
| 11 |
|
| 12 |
-
|
| 13 |
-
- Classify flowers into 5 species
|
| 14 |
-
- For educational and research use
|
| 15 |
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
-
|
| 20 |
|
| 21 |
-
##
|
| 22 |
-
- Conv Layers: 16 → 32 → 64 → 128 filters
|
| 23 |
-
- Dense Layer: 128 units + output layer
|
| 24 |
-
- Activation: ReLU (Conv/Dense), Softmax (Output)
|
| 25 |
-
- Loss: categorical_crossentropy
|
| 26 |
-
- Optimizer: Adam
|
| 27 |
-
- Dropout: 0.5
|
| 28 |
|
| 29 |
-
|
| 30 |
-
-
|
| 31 |
-
-
|
| 32 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
## Performance
|
| 35 |
-
- Training Accuracy: 90%
|
| 36 |
-
- Validation Accuracy: 80%
|
| 37 |
-
- Validation Loss: [mention latest]
|
| 38 |
|
| 39 |
-
|
| 40 |
-
-
|
| 41 |
-
-
|
| 42 |
-
- May fail on noisy/real-world images
|
| 43 |
|
| 44 |
## Usage
|
| 45 |
-
- Predict flower species from image
|
| 46 |
-
- Research, educational, hobby projects
|
| 47 |
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language: en
|
| 3 |
+
tags:
|
| 4 |
+
- computer-vision
|
| 5 |
+
- image-classification
|
| 6 |
+
- flowers
|
| 7 |
+
- plants
|
| 8 |
+
- biology
|
| 9 |
+
- tensorflow
|
| 10 |
+
- efficientnet
|
| 11 |
+
library_name: tensorflow
|
| 12 |
+
datasets:
|
| 13 |
+
- flowers-recognition
|
| 14 |
license: mit
|
| 15 |
---
|
|
|
|
| 16 |
|
| 17 |
+
# Plant Species Classification Model
|
|
|
|
| 18 |
|
| 19 |
+
## Model Description
|
|
|
|
| 20 |
|
| 21 |
+
This is a deep learning model for automated classification of flower species using computer vision. The model can identify 5 common flower types with high accuracy.
|
|
|
|
|
|
|
| 22 |
|
| 23 |
+
**Model type:** Image Classification
|
| 24 |
+
**Architecture:** EfficientNetB0 with custom classification head
|
| 25 |
+
**Input:** 224×224 RGB images
|
| 26 |
+
**Output:** 5-class classification probabilities
|
| 27 |
|
| 28 |
+
## Intended Uses
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
+
### Primary Use Cases
|
| 31 |
+
- 🌿 Educational plant identification tools
|
| 32 |
+
- 📱 Mobile flower recognition applications
|
| 33 |
+
- 🔬 Botanical research and biodiversity monitoring
|
| 34 |
+
- 🌸 Gardening and nature enthusiast apps
|
| 35 |
+
|
| 36 |
+
### Limitations
|
| 37 |
+
- Trained on only 5 specific flower species
|
| 38 |
+
- Performance may vary with image quality and lighting conditions
|
| 39 |
+
- Not suitable for rare or unusual flower varieties
|
| 40 |
+
|
| 41 |
+
## Classes
|
| 42 |
+
|
| 43 |
+
The model classifies images into 5 flower species:
|
| 44 |
+
|
| 45 |
+
1. **daisy** 🌼 - Classic white petals with yellow center
|
| 46 |
+
2. **dandelion** 🌞 - Bright yellow composite flowers
|
| 47 |
+
3. **rose** 🌹 - Layered petals in various colors
|
| 48 |
+
4. **sunflower** 🌻 - Large yellow flowers with dark centers
|
| 49 |
+
5. **tulip** 🌷 - Cup-shaped flowers in vibrant colors
|
| 50 |
+
|
| 51 |
+
## Training Data
|
| 52 |
+
|
| 53 |
+
- **Dataset:** Flowers Recognition from Kaggle
|
| 54 |
+
- **Total Images:** ~4,300
|
| 55 |
+
- **Split:** 80% training, 20% validation
|
| 56 |
+
- **Augmentation:** Rotation, flipping, zooming, brightness adjustment
|
| 57 |
|
| 58 |
## Performance
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
+
- **Validation Accuracy:** >90%
|
| 61 |
+
- **Inference Speed:** Real-time capable
|
| 62 |
+
- **Model Size:** ~30MB
|
|
|
|
| 63 |
|
| 64 |
## Usage
|
|
|
|
|
|
|
| 65 |
|
| 66 |
+
```python
|
| 67 |
+
from tensorflow.keras.models import load_model
|
| 68 |
+
from tensorflow.keras.preprocessing import image
|
| 69 |
+
import numpy as np
|
| 70 |
+
|
| 71 |
+
# Load and use the model for flower classification
|
| 72 |
+
model = load_model('flower_classification_model.h5')
|
| 73 |
+
```
|
| 74 |
+
|
| 75 |
+
**Input Requirements:**
|
| 76 |
+
- Image format: JPEG, PNG
|
| 77 |
+
- Image size: 224×224 pixels
|
| 78 |
+
- Color mode: RGB
|
| 79 |
+
|
| 80 |
+
## Ethical Considerations
|
| 81 |
+
|
| 82 |
+
- Intended for educational and research purposes
|
| 83 |
+
- Should not replace expert botanical identification
|
| 84 |
+
- Respect privacy when deploying in applications
|
| 85 |
+
|
| 86 |
+
## Citation
|
| 87 |
+
|
| 88 |
+
If you use this model in your work, please cite:
|
| 89 |
+
```
|
| 90 |
+
Plant Species Classification Model by Athar Abbas
|
| 91 |
+
https://huggingface.co/AtharAbbas993/Plant_Species_Classification
|
| 92 |
+
```
|