app.py

import os
import tempfile
import base64
import cv2
import streamlit as st
import PIL
from ultralytics import YOLO
import requests

###############################################################################
# Helper function to embed an HTML5 video that autoplays (muted) with controls.
###############################################################################
def show_autoplay_video(video_bytes: bytes, title: str = "Video"):
    if not video_bytes:
        st.warning(f"No {title} video available.")
        return
    video_base64 = base64.b64encode(video_bytes).decode()
    video_html = f"""
    <h4>{title}</h4>
    <video width="100%" controls autoplay muted>
      <source src="data:video/mp4;base64,{video_base64}" type="video/mp4">
      Your browser does not support the video tag.
    </video>
    """
    st.markdown(video_html, unsafe_allow_html=True)

###############################################################################
# Session state initialization (for uploaded processing results)
###############################################################################
if "processed_frames" not in st.session_state:
    st.session_state["processed_frames"] = []
if "shortened_video_data" not in st.session_state:
    st.session_state["shortened_video_data"] = None
if "shortened_video_ready" not in st.session_state:
    st.session_state["shortened_video_ready"] = False

###############################################################################
# Configure YOLO model path and page layout
###############################################################################
model_path = 'https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/best.pt'
st.set_page_config(
    page_title="Fire Detection: Original vs. Processed Video",
    page_icon="🔥",
    layout="wide",
    initial_sidebar_state="expanded"
)

st.title("Fire Watch: Detecting fire using AI vision models")
col1, col2 = st.columns(2)
with col1:
    st.image("https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/Fire_1.jpeg", use_column_width=True)
with col2:
    st.image("https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/Fire_3.png", use_column_width=True)

st.markdown("""
Fires in Colorado present a serious challenge, threatening urban communities, highways, and even remote areas. 
Early detection is critical. Fire Watch uses the model YOLOv8 for real-time fire and smoke detection 
in images and videos.
""")
st.markdown("---")
st.header("Fire Detection:")


###############################################################################
# SIDEBAR: Video input options, confidence, sampling options, and example selection
###############################################################################
with st.sidebar:
    st.header("Video Input Options")
    # Option to select an example pair; "None" means use an uploaded file.
    example_option = st.selectbox(
        "Select Example Pair (optional)",
        ["None", "T Example", "LA Example"]
    )
    source_file = st.file_uploader(
        "Or upload your own file...",
        type=("mp4", "jpg", "jpeg", "png", "bmp", "webp")
    )
    confidence = float(st.slider("Select Model Confidence", 25, 100, 40)) / 100
    video_option = st.selectbox(
        "Select Video Shortening Option",
        ["Original FPS", "1 fps", "1 frame per 5 seconds", "1 frame per 10 seconds", "1 frame per 15 seconds"]
    )
    progress_text = st.empty()
    progress_bar = st.progress(0)

###############################################################################
# MAIN TITLE
###############################################################################
st.title("Fire Detection: Original vs. Processed Video")

###############################################################################
# Load YOLO model
###############################################################################
try:
    model = YOLO(model_path)
except Exception as ex:
    st.error(f"Unable to load model. Check model path: {model_path}")
    st.error(ex)

###############################################################################
# Determine source video(s): Example pair or uploaded file.
###############################################################################
original_video_data = None
processed_video_data = None  # For example pairs

if example_option != "None":
    # Use example videos from remote URLs.
    if example_option == "T Example":
        # For T Example: set your URLs for original and processed videos.
        orig_url = "https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/T1.mp4"
        proc_url = "https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/T2.mpg"
    elif example_option == "LA Example":
        orig_url = "https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/LA1.mp4"
        proc_url = "https://huggingface.co/spaces/tstone87/ccr-colorado/resolve/main/LA2.mp4"
    try:
        original_video_data = requests.get(orig_url).content
        processed_video_data = requests.get(proc_url).content
    except Exception as ex:
        st.error("Error loading example videos. Check your URLs.")
else:
    # No example selected. If a file is uploaded, use it.
    if source_file:
        file_type = source_file.type.split('/')[0]
        if file_type == 'image':
            # For images, convert to video-like display (or you could run image detection).
            original_image = PIL.Image.open(source_file)
            buf = tempfile.NamedTemporaryFile(suffix=".png", delete=False)
            original_image.save(buf.name, format="PNG")
            with open(buf.name, "rb") as f:
                original_video_data = f.read()
        else:
            # For video uploads, save to a temp file.
            tfile = tempfile.NamedTemporaryFile(delete=False, suffix=".mp4")
            tfile.write(source_file.read())
            tfile.flush()
            with open(tfile.name, "rb") as vf:
                original_video_data = vf.read()
            # Open with OpenCV for processing.
            vidcap = cv2.VideoCapture(tfile.name)
    else:
        st.info("Please select an example pair or upload a video file.")

###############################################################################
# Layout: Two columns for Original and Processed videos
###############################################################################
col1, col2 = st.columns(2)

with col1:
    st.subheader("Original File")
    if original_video_data:
        show_autoplay_video(original_video_data, title="Original Video")
    else:
        st.info("No original video available.")

with col2:
    st.subheader("Result File")
    if example_option != "None":
        # For example pairs, the processed video is already available.
        if processed_video_data:
            show_autoplay_video(processed_video_data, title="Processed Video")
        else:
            st.info("No processed video available in example.")
    else:
        # For uploaded files, if a processed video is ready, show it.
        if st.session_state["shortened_video_ready"] and st.session_state["shortened_video_data"]:
            show_autoplay_video(st.session_state["shortened_video_data"], title="Processed Video")
        else:
            st.info("Processed video will appear here once detection is run.")

###############################################################################
# DETECTION: Process the uploaded video if no example is selected.
###############################################################################
if example_option == "None" and source_file and source_file.type.split('/')[0] != 'image':
    if st.sidebar.button("Let's Detect Wildfire"):
        # Reset any previous processed results.
        st.session_state["processed_frames"] = []
        st.session_state["shortened_video_data"] = None
        st.session_state["shortened_video_ready"] = False

        processed_frames = st.session_state["processed_frames"]

        frame_count = 0
        orig_fps = vidcap.get(cv2.CAP_PROP_FPS)
        total_frames = int(vidcap.get(cv2.CAP_PROP_FRAME_COUNT))
        width  = int(vidcap.get(cv2.CAP_PROP_FRAME_WIDTH))
        height = int(vidcap.get(cv2.CAP_PROP_FRAME_HEIGHT))

        # Determine sampling interval based on option.
        if video_option == "Original FPS":
            sample_interval = 1
            output_fps = orig_fps
        elif video_option == "1 fps":
            sample_interval = int(orig_fps) if orig_fps > 0 else 1
            output_fps = 1
        elif video_option == "1 frame per 5 seconds":
            sample_interval = int(orig_fps * 5) if orig_fps > 0 else 5
            output_fps = 1
        elif video_option == "1 frame per 10 seconds":
            sample_interval = int(orig_fps * 10) if orig_fps > 0 else 10
            output_fps = 1
        elif video_option == "1 frame per 15 seconds":
            sample_interval = int(orig_fps * 15) if orig_fps > 0 else 15
            output_fps = 1
        else:
            sample_interval = 1
            output_fps = orig_fps

        success, image = vidcap.read()
        while success:
            if frame_count % sample_interval == 0:
                res = model.predict(image, conf=confidence)
                res_plotted = res[0].plot()[:, :, ::-1]
                processed_frames.append(res_plotted)
                # Update progress
                if total_frames > 0:
                    progress_pct = int((frame_count / total_frames) * 100)
                    progress_text.text(f"Processing frame {frame_count} / {total_frames} ({progress_pct}%)")
                    progress_bar.progress(min(100, progress_pct))
                else:
                    progress_text.text(f"Processing frame {frame_count}")
            frame_count += 1
            success, image = vidcap.read()

        progress_text.text("Video processing complete!")
        progress_bar.progress(100)

        # Create shortened video from processed frames.
        if processed_frames:
            temp_video_file = tempfile.NamedTemporaryFile(delete=False, suffix='.mp4')
            fourcc = cv2.VideoWriter_fourcc(*'mp4v')
            out = cv2.VideoWriter(temp_video_file.name, fourcc, output_fps, (width, height))
            for frame in processed_frames:
                out.write(frame)
            out.release()

            with open(temp_video_file.name, 'rb') as video_file:
                st.session_state["shortened_video_data"] = video_file.read()
                st.session_state["shortened_video_ready"] = True

            st.success("Processed video created successfully!")
        else:
            st.error("No frames were processed from the video.")

###############################################################################
# Always show the download button if a processed video is ready.
###############################################################################
if st.session_state["shortened_video_ready"] and st.session_state["shortened_video_data"]:
    st.download_button(
        label="Download Processed Video",
        data=st.session_state["shortened_video_data"],
        file_name="processed_video.mp4",
        mime="video/mp4"
    )