Khalid-Hamad/plant-seedlings-classifier

Plant Seedlings Classifier

A CNN model that classifies 12 species of plant seedlings, trained on the Plant Seedlings Dataset.

Species

Black-grass	Charlock	Cleavers
Common Chickweed	Common wheat	Fat Hen
Loose Silky-bent	Maize	Scentless Mayweed
Shepherds Purse	Small-flowered Cranesbill	Sugar beet

Architecture

Conv2D(32, 5×5) → BatchNorm → MaxPool → Dropout(0.2)
Conv2D(64, 5×5) → BatchNorm → MaxPool → Dropout(0.3)
Conv2D(64, 3×3) → BatchNorm → MaxPool → Dropout(0.4)
Conv2D(64, 3×3) → BatchNorm → MaxPool → Dropout(0.5)
GlobalMaxPooling2D → Dense(256) → Dropout(0.5) → Dense(12, softmax)

• Input: 128×128×3 RGB images
• Optimizer: Adam with ReduceLROnPlateau
• Data augmentation: rotation, shift, flip, zoom
• Early stopping with best weight restoration

Usage

import numpy as np
import tensorflow as tf

model = tf.keras.models.load_model("plant_seedling_model.keras")
class_names = np.load("class_names.npy", allow_pickle=True)

Files

• plant_seedling_model.keras — Trained Keras model
• class_names.npy — Array of 12 class label strings

Training

Trained on Google Colab with GPU. See the full training notebook at github.com/Khalid-Hamad/seedling-species-cv-cnn.

License

Model weights: educational use. Dataset: CC BY-SA 4.0.