import streamlit as st
import cv2
import numpy as np
import tensorflow as tf
import random
import time

url = st.text_area('enter streaming url')
locations = ['Miami', 'Smouha', 'Mandara', 'Sporting', 'Montazah']
model = tf.saved_model.load('my_model')

classes_1 = ["Arson", "Assault", "Burglary","Normal"]  # Define your classes here

# Function to preprocess frames for action recognition
def preprocess_frame(frame):
    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    frame = cv2.resize(frame, (224, 224))
    frame = frame.astype('float32') / 255.0
    frame = np.expand_dims(frame, axis=0)
    return frame

def get_top_k(probs, k=1, label_map=classes_1):
    """Outputs the top k model labels and probabilities on the given video."""
    top_predictions = tf.argsort(probs, axis=-1, direction='DESCENDING')[:k]
    top_labels = tf.gather(label_map, top_predictions, axis=-1)
    top_labels = [label.decode('utf8') for label in top_labels.numpy()]
    top_probs = tf.gather(probs, top_predictions, axis=-1).numpy()
    return top_labels[0]

# Function to perform action recognition on the HLS stream
def perform_action_recognition(url, model, k=1, label_map=classes_1, locations=locations):
    # Open the video stream
    cap = cv2.VideoCapture(url)
    start_time = time.time()  # Capture the start time

    while True:
        # Read the next frame
        ret, frame = cap.read()

        if not ret:
            break

        # Preprocess the frame
        preprocessed_frame = preprocess_frame(frame)

        # Perform action recognition on the preprocessed frame
        outputs = model.signatures['serving_default'](image=preprocessed_frame[tf.newaxis])
        probs = tf.nn.softmax(outputs['classifier_head_1'])
        current_time = time.time() - start_time  # Calculate the elapsed time
        m, s = divmod(current_time, 60)
        h, m = divmod(m, 60)
        ip_address = last_part = url.split("/")[-1]
        output = [get_top_k(probs[0], k=k, label_map=label_map), f"{int(h):02d}:{int(m):02d}:{int(s):02d}", random.choice(locations), ip_address]
        yield output

    # Release the video stream
    cap.release()

# Call the function to perform action recognition on the HLS stream

# Create a placeholder for the output
output_placeholder = st.empty()

# Call the function to perform action recognition on the HLS stream and update the placeholder
if url:
    for output in perform_action_recognition(url, model):
        output_placeholder.json(output)