app.py

import streamlit as st
from tensorflow.keras.models import load_model
from PIL import Image
import cv2
import numpy as np

def resize_image(image):
    '''
    Function to resize an image into 128 * 128
    '''
    resized_image = cv2.resize(image, (128, 128), interpolation=cv2.INTER_AREA)
    return resized_image

def predict(image):
    # Loading model
    model = load_model('InceptionV3.h5')

    # Resize image into 128x128
    resized_image = resize_image(image)

    # Convert into numpy array
    image_array = np.array(resized_image)

    # Rescale image (0-1)
    rescaled_image = image_array.astype(np.float32) / 255.0

    # Expanding dimensions
    input_image = np.expand_dims(rescaled_image, axis=0)

    # Make prediction
    pred = model.predict(input_image)

    # Convert the prediction to a label
    if pred[0] > 0.5:
        label = 'Parasitized Cell'
    else:
        label = 'Uninfected Cell'
    return label

def main():
    st.title("Malaria Disease Detection")
    st.write("Upload an image of a cell, and we will predict if it is parasitized or uninfected.")

    # File uploader
    uploaded_file = st.file_uploader("Choose an image...", type=['jpg', 'jpeg', 'png'])

    if uploaded_file is not None:
        # Read image
        image = Image.open(uploaded_file)

        # Convert PIL Image to numpy array
        image_array = np.array(image)

        # Make prediction
        prediction = predict(image_array)
        prediction = "<h3 style='font-family: Arial;'>Prediction: " + prediction + "</h3>"
        st.write(prediction, unsafe_allow_html=True)

if __name__ == '__main__':
    main()