waste_classification.py

# -*- coding: utf-8 -*-
"""Waste classification.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1He-SjfOmrPOdw7HiyFGKwzcZhWiBbAgx

####**Waste** **Classification** **using** **CNN**
***Introduction***

* Waste is a significant global issue. Increasing volumes of waste are being generated as the global population and living standards rise. People are increasingly concerned about the production of waste and its effect, and are seeking ways to deal with the problem.

* Recycling is the process of converting waste materials into new materials and objects. The recovery of energy from waste materials is often included in this concept. The recyclability of a material depends on its ability to reacquire the properties it had in its original state. It is an alternative to "conventional" waste disposal that can save material and help lower greenhouse gas emissions. Recycling can prevent the waste of potentially useful materials and reduce the consumption of fresh raw materials, thereby reducing: energy usage, air pollution (from incineration), and water pollution (from landfilling)


🟢 In this notebook, we will classify waste as organic or recyclable using Convolutional Neural Network (CNN).
"""

from google.colab import drive
drive.mount('/content/drive')

# Paths to training and testing datasets in Drive
train_path = "/content/drive/MyDrive/DATASET/TRAIN/"
test_path  = "/content/drive/MyDrive/DATASET/TEST/"

import os
import matplotlib.pyplot as plt
import cv2

# Path to your training dataset
train_path = '/content/drive/MyDrive/DATASET/TRAIN/'

# Classes
classes = ['O', 'R']

for cls in classes:
    class_path = os.path.join(train_path, cls)
    if os.path.exists(class_path):
        print(f"\nClass: {cls}")
        images = os.listdir(class_path)

        # Display all images one by one
        for img_name in images:
            img_path = os.path.join(class_path, img_name)
            img = cv2.imread(img_path)
            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

            plt.imshow(img)
            plt.title(f"Class: {cls} | File: {img_name}")
            plt.axis("off")
            plt.show()
    else:
        print(f"Class {cls} folder not found!")

"""#####Import Libraries

"""

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Activation, Dropout, Flatten, Dense, BatchNormalization
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.utils import plot_model
from tensorflow.keras.utils import img_to_array, load_img   # now comes from utils
from glob import glob

"""###Visualization"""

from collections import Counter
import pandas as pd
import matplotlib.pyplot as plt

label_counts = Counter(y_data)

df_counts = pd.DataFrame.from_dict(label_counts, orient='index', columns=['count']).reset_index()
df_counts.rename(columns={'index': 'label'}, inplace=True)

plt.figure(figsize=(8,5))
bars = plt.bar(df_counts['label'], df_counts['count'], color='skyblue', edgecolor='black')
plt.title("Class Distribution in Dataset", fontsize=14)
plt.xlabel("Class Labels", fontsize=12)
plt.ylabel("Number of Images", fontsize=12)
plt.xticks(rotation=45)

# Add counts on top of bars
for bar, count in zip(bars, df_counts['count']):
    plt.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 5,
             str(count), ha='center', va='bottom')

plt.show()

print("x_data shape:", x_data.shape)   # e.g. (num_images, 128, 128, 3)
print("y_data shape:", y_data.shape)   # e.g. (num_images,)

from collections import Counter
Counter(y_data)

import matplotlib.pyplot as plt
from collections import Counter

# Example dataset labels
y_data = ['Organic', 'Recyclable', 'Organic', 'Organic', 'Recyclable', 'Organic']

# Custom colors (green for Organic, pink for Recyclable)
colors = ['#a0d157', '#c48bb8']

# Count labels
label_counts = Counter(y_data)

# Pie chart
plt.figure(figsize=(6,6))
plt.pie(
    label_counts.values(),
    startangle=90,
    explode=[0.05]*len(label_counts),     # explode all slices slightly
    autopct='%0.2f%%',                    # show percentages with 2 decimals
    labels=label_counts.keys(),
    colors=colors[:len(label_counts)],    # assign colors dynamically
    radius=1.2
)

plt.title("Class Distribution in Dataset", fontsize=14)
plt.show()

print(set(y_data))

import os

train_path = "/content/drive/MyDrive/DATASET/TRAIN/"

# Show classes
classes = os.listdir(train_path)
print("Classes found:", classes)

# Show file counts per class
for cls in classes:
    folder = os.path.join(train_path, cls)
    files = os.listdir(folder)
    print(f"Class: {cls}, Files: {len(files)}")
    print("Sample files:", files[:5])  # show first 5

className = glob(train_path + '/*' )
numberOfClass = len(className)
print("Number Of Class: ",numberOfClass)

"""######Convolutional Neural Network - CNN"""

model = Sequential()
model.add(Conv2D(32,(3,3),input_shape=(224,224,3)))
model.add(Activation("relu"))
model.add(MaxPooling2D())

model.add(Conv2D(64,(3,3)))
model.add(Activation("relu"))
model.add(MaxPooling2D())

model.add(Conv2D(128,(3,3)))
model.add(Activation("relu"))
model.add(MaxPooling2D())

model.add(Flatten())
model.add(Dense(256))
model.add(Activation("relu"))
model.add(Dropout(0.5))
model.add(Dense(64))
model.add(Activation("relu"))
model.add(Dropout(0.5))
model.add(Dense(numberOfClass, activation="softmax"))   # ✅ Softmax

model.compile(loss="categorical_crossentropy",   # ✅
              optimizer="adam",
              metrics=["accuracy"])

batch_size = 32   # ✅ safer value

plot_model(model)

train_datagen = ImageDataGenerator(rescale= 1./255)

test_datagen = ImageDataGenerator(rescale= 1./255)

train_generator = train_datagen.flow_from_directory(
        train_path,
        target_size= (224,224),
        batch_size = batch_size,
        color_mode= "rgb",
        class_mode= "categorical")

test_generator = test_datagen.flow_from_directory(
        test_path,
        target_size= (224,224),
        batch_size = batch_size,
        color_mode= "rgb",
        class_mode= "categorical")

hist = model.fit(
        train_generator,
        epochs=10,
        validation_data=test_generator)

plt.figure(figsize=[10,6])
plt.plot(hist.history["accuracy"], label="Train Accuracy")
plt.plot(hist.history["val_accuracy"], label="Validation Accuracy")
plt.title("Training vs Validation Accuracy")
plt.xlabel("Epochs")
plt.ylabel("Accuracy")
plt.legend()
plt.grid(True)
plt.show()

plt.figure(figsize=(10,6))
plt.plot(hist.history['loss'], label = "Train loss")
plt.plot(hist.history['val_loss'], label = "Validation loss")
plt.legend()
plt.show()

"""##Model **Prediction**


"""

import cv2
import matplotlib.pyplot as plt
import numpy as np

def predict_func(img):
    # Show the input image
    plt.figure(figsize=(6,4))
    plt.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
    plt.axis("off")
    plt.tight_layout()
    plt.show()

    # Preprocess for model
    img_resized = cv2.resize(img, (224, 224))
    img_resized = np.reshape(img_resized, [1, 224, 224, 3])  # batch size = 1

    # Prediction
    result = np.argmax(model.predict(img_resized))

    # Output class
    if result == 0:
        print("\033[94m" + "This image -> Recyclable" + "\033[0m")
    elif result == 1:
        print("\033[94m" + "This image -> Organic" + "\033[0m")
    else:
        print("\033[91m" + "Unknown class" + "\033[0m")


# Load test image from Google Drive
test_img = cv2.imread("/content/drive/MyDrive/DATASET/TEST/O/O_12573.jpg")

if test_img is None:
    print("Error: Image not found. Please check the path.")
else:
    predict_func(test_img)

import cv2

# Now your code works
test_img = cv2.imread("/content/drive/MyDrive/DATASET/TEST/O/O_13487.jpg")

if test_img is None:
    print("Error: Image not found. Please check the path.")
else:
    predict_func(test_img)

def predict_func(img):
    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img = cv2.resize(img, (224, 224))
    img = np.expand_dims(img, axis=0) / 255.0
    return model.predict(img)

# Now call it
test_img = cv2.imread("/content/drive/MyDrive/DATASET/TEST/R/R_10753.jpg")
print(predict_func(test_img))