app.py

import streamlit as st
import onnxruntime as ort
import cv2
import numpy as np
from PIL import Image

# Set page configuration
st.set_page_config(page_title="Emotion Recognition App", layout="centered")
st.title("Emotion Recognition App")

# Load the ONNX model using onnxruntime
@st.cache_resource
def load_model():
    model_path = "onnx_model.onnx"  # Ensure this is the correct path to your uploaded ONNX model
    return ort.InferenceSession(model_path)

# Load the emotion detection model
emotion_model = load_model()

# Process the uploaded image
uploaded_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])

def preprocess_image(image):
    """Preprocess image to match model input requirements"""
    image = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2GRAY)  # Convert image to grayscale
    image_resized = cv2.resize(image, (48, 48))  # Resize image to 48x48
    image_input = np.expand_dims(image_resized, axis=0)  # Add batch dimension
    image_input = np.expand_dims(image_input, axis=0)  # Add channel dimension (for grayscale)
    image_input = image_input.astype(np.float32) / 255.0  # Normalize the image
    return image_input

def predict_emotion(image_input):
    """Run inference and predict the emotion"""
    input_name = emotion_model.get_inputs()[0].name
    output_name = emotion_model.get_outputs()[0].name
    prediction = emotion_model.run([output_name], {input_name: image_input})
    emotion = np.argmax(prediction[0])  # Get the class with the highest probability
    return emotion

# Define a function to display emotion text
def display_emotion(emotion):
    """Map emotion index to a human-readable emotion"""
    emotion_map = {
        0: "Anger",
        1: "Disgust",
        2: "Fear",
        3: "Happiness",
        4: "Sadness",
        5: "Surprise",
        6: "Neutral"
    }
    return emotion_map.get(emotion, "Unknown")

# If an image is uploaded
if uploaded_file is not None:
    # Open and display the uploaded image
    image = Image.open(uploaded_file)
    st.image(image, caption="Uploaded Image", use_column_width=True)

    # Preprocess the image
    image_input = preprocess_image(image)

    # Predict the emotion
    emotion = predict_emotion(image_input)
    emotion_label = display_emotion(emotion)

    # Display the predicted emotion
    st.write(f"Detected Emotion: {emotion_label}")