Upload 2 files

Home.py

@@ -0,0 +1,62 @@
+import joblib
+import streamlit as st
+from prediction import  predict_single_image
+
+knn_model = joblib.load('models/knn_model.joblib')
+svm_model = joblib.load('models/svm_model.joblib')
+random_forest_model = joblib.load('models/random_forest_model.joblib')
+
+def show_error_popup(message):
+    st.error(message, icon="🚨")
+
+st.set_page_config(layout="wide")
+
+st.title('CASIA PALMPRINT DATASET')
+st.markdown('By Yash Patel')
+
+st.header('Add Palmprint Image')
+
+uploaded_file = st.file_uploader("Choose an image", type=["jpg", "png", "jpeg"])
+
+st.header("Available Models")
+option = st.selectbox(
+    "Available Models",
+    ("SVM", "KNN","Random Forest"),
+)
+
+predicted_label =""
+col1, col2= st.columns(2)
+
+if uploaded_file is not None:
+    with col1:
+        image_data = uploaded_file.read()
+        st.image(image_data, caption="Uploaded Image")
+    with col2:
+        if option=="SVM":
+            predicted_label = predict_single_image(svm_model,image_data)
+        elif option=="KNN":
+            predicted_label = predict_single_image(knn_model, image_data)
+        elif option=="Random Forest":
+            predicted_label = predict_single_image(random_forest_model, image_data)
+        else:
+            p = "Other Models are still under training due to overfitting"
+
+        print(predicted_label)
+
+        st.markdown("""
+        <style>
+        .big-font {
+            display: flex;
+            align-items:center;
+            justify-content: center;
+            font-size:50px !important;
+            color:green;
+            height: 50vh;
+        }
+        </style>
+        """, unsafe_allow_html=True)
+
+        st.markdown(f'<div class="big-font">{predicted_label}</div>', unsafe_allow_html=True)
+
+else:
+    show_error_popup("Please Upload Image...")

prediction.py

@@ -0,0 +1,28 @@
+import cv2
+import numpy as np
+import tensorflow as tf
+
+def preprocess_image(img):
+    """Preprocess a single image for prediction."""
+    img = tf.image.decode_jpeg(img, channels=1)
+    img= tf.image.resize(img, (224, 224))
+    img_flattened = tf.reshape(img, (-1,))
+
+    # Convert to 2D array (expected input format for the model)
+    img_flattened = np.expand_dims(img_flattened, axis=0)  # Shape: (1, features)
+
+    return img_flattened
+
+
+def predict_single_image(model, image):
+    """Predict the label of a single image."""
+    # Preprocess the image
+    processed_image = preprocess_image(image)
+
+    # Make prediction
+    prediction = model.predict(processed_image)
+
+    return prediction[0]  # Return the predicted label
+
+
+# Test the single image