diff --git a/app.py b/app.py
index 28e29b5..445edb9 100644
--- a/app.py
+++ b/app.py
@@ -1,53 +1,38 @@
 import streamlit as st
-from PIL import Image
-from app_module import classify
+import numpy as np
+from tensorflow.keras.models import model_from_json
+from tensorflow.keras.preprocessing import image
+from tensorflow.keras.applications.imagenet_utils import preprocess_input
 
-st.set_page_config(page_title='Solar Photovoltaics Checker', page_icon='model/solar_logo.jpeg', layout="centered", initial_sidebar_state="auto", menu_items=None)
+# Load the model architecture from JSON
+with open('D:/Projects/solar-panel-dust-detection/Dust-Detection/model/best_model.json', 'r') as json_file:
+    model_json = json_file.read()
 
+# Convert JSON string to model
+model = model_from_json(model_json)
 
-st.image('ricesurelogo.jpeg', width=200)  
-tab1, tab2, tab3 = st.tabs(["Prediction and Quality Check", "Solar Panel Information", "Student Information"])
+# Load the model weights
+model.load_weights('D:/Projects/solar-panel-dust-detection/Dust-Detection/model/best_model.h5')
 
-with tab1:
-        with st.form:
-        #st.title("Rice Sure")
-        #st.divider()
-        st.subheader("RiceSure app fights fraudulent in the rice industry by using Convolutional Neural Network with Transfer Learning and Fine Tuning to identify rice grain types and ensure consistent quality control.")
-        st.subheader("Some of the rice varieties including Basmati Rice, Jasmine Rice, Karacadag Rice, Arborio Rice, and Ipsala have different features and structure. Let's predict the type of the rice below. ")
-        #st.divider()
-        #st.subheader("Upload rice grain image.")
-        uploaded_file = st.file_uploader("Choose an image of a rice grain. (Kindly attach only rice grain images)")
-        if uploaded_file is not None:
-                image = Image.open(uploaded_file)
-                st.image(image, caption='Uploaded Image')
-                image.save('up.jpg')
-                #st.divider()
-                
-                if st.button('Click to process'):
-                        st.spinner(text="Predicting...")
-                        #st.balloons()
-                        st.success('The image was uploaded successfully and below are the rice type and rice quality check')
-                        label = classify(uploaded_file)
-                        quality = rice_quality_check('up.jpg')
-                        
-                        #res = sign_names.get(label)
-                        st.markdown(label)
-                        st.markdown(quality)
-with tab2:
-        st.subheader("Basmati Rice")
-        st.image("Basmati.jpg", width=200)
-        st.write("Basmati rice is a long-grain rice variety grown primarily in India and Pakistan. It has a distinctive nutty flavor and aroma, and the cooked grains are fluffy and separate, making it a popular choice for biryanis, pilafs, and other Indian and Middle Eastern dishes.")
-        #st.divider()
-        
-        st.subheader("Jasmine Rice")
-        st.image("Jasmine.jpg", width=200)
-        st.write("Jasmine rice is a long-grain rice variety grown in Thailand, Cambodia, and Vietnam. It has a fragrant aroma and a slightly sweet flavor, making it popular in Asian cuisine. Jasmine rice is fluffy and soft when cooked and is often served alongside curries, stir-fries, and other spicy dishes.")
-        #st.divider()
+# Streamlit UI
+st.title("Solar Panel Dust Detection")
 
-with tab3:
-        st.write("Rajasegaran a/l M Sivaanandan (22052733)")
-        st.write("Basubeit, Omar Gumaan Saleh (22054606)")
-        st.write("Safwan bin Shamsir (S2195293)")
-        st.write("Azle Abd Ghalim (S2036517)")
-        st.write("Intan Nor Qamarina Binti Yunus (S2180929)")
+# Upload image
+uploaded_file = st.file_uploader("Choose an image...", type="jpg")
 
+if uploaded_file is not None:
+    # Display image
+    img = image.load_img(uploaded_file, target_size=(224, 224))
+    st.image(img, caption='Uploaded Image', use_column_width=True)
+    st.write("")
+    
+    # Preprocess image
+    img_array = image.img_to_array(img)
+    img_array = np.expand_dims(img_array, axis=0)
+    img_array = preprocess_input(img_array)
+
+    # Predict
+    prediction = model.predict(img_array)
+    
+    # Display prediction result
+    st.write("Prediction:", prediction)