app.py

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

model=load_model('pneumonia.h5')

# def process_image(img):
#     img=img.resize((50,50)) #boyutunu 50 x 50 pixel yaptik
#     img=np.array(img)
#     img=img/255.0 #normalize ettik
#     img=np.expand_dims(img,axis=0)
#     return img

def process_image(img):
    # Resize the image while maintaining the aspect ratio
    img = ImageOps.fit(img, (50, 50), Image.BICUBIC, centering=(0.5, 0.5))
    
    # Convert the image to a numpy array and add the channel dimension if needed
    img = np.array(img)
    if img.ndim == 2:
        img = np.expand_dims(img, axis=-1)
        img = np.repeat(img, 3, axis=-1)
    elif img.ndim == 3 and img.shape[-1] == 1:
        img = np.repeat(img, 3, axis=-1)
    
    # Normalize the pixel values
    img = img / 255.0
    
    # Add the batch dimension
    img = np.expand_dims(img, axis=0)
    
    return img

st.title("Pnömoni tespiti :lungs:")
st.write("Pnömoni olup olmadığını öğrenmek için röntgen filmi resmini yükle")

file=st.file_uploader('Bir Resim Sec',type=['jpg','jpeg','png'])

if file is not None:
    img=Image.open(file)
    st.image(img,caption='yuklenen resim')
    image= process_image(img)
    prediction=model.predict(image)
    predicted_class=np.argmax(prediction)
    
    class_names=['normal','pneumonia']
    st.write(class_names[predicted_class])