app2.py

import streamlit as st
from PIL import Image
from ultralytics import YOLO
import cv2, os
from datetime import datetime
import numpy as np

# ---------------- Config générale ----------------
MODEL_PATH = "best.pt"
SAVE_DIR = os.path.join("/tmp", "results")
os.makedirs(SAVE_DIR, exist_ok=True)

# Charger le modèle
model = YOLO(MODEL_PATH)

# ---------------- Fonctions utilitaires ----------------
def predict_image(image, conf=0.25, show_labels=True):
    image = np.array(image)
    if image.shape[2] == 4:
        image = cv2.cvtColor(image, cv2.COLOR_RGBA2BGR)
    else:
        image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)

    results = model.predict(source=image, conf=conf, verbose=False)
    annotated_image = results[0].plot(labels=show_labels)

    out_path = os.path.join(SAVE_DIR, f"image_result_{datetime.now().strftime('%Y%m%d_%H%M%S')}.png")
    cv2.imwrite(out_path, annotated_image)
    return out_path

def predict_video(video_path, conf=0.25, show_labels=True):
    cap = cv2.VideoCapture(video_path)
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out_path = os.path.join(SAVE_DIR, f"video_result_{datetime.now().strftime('%Y%m%d_%H%M%S')}.mp4")

    fps = cap.get(cv2.CAP_PROP_FPS) or 30
    width, height = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)), int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
    out = cv2.VideoWriter(out_path, fourcc, fps, (width, height))

    while cap.isOpened():
        ret, frame = cap.read()
        if not ret:
            break
        results = model.predict(frame, conf=conf, verbose=False)
        annotated = results[0].plot(labels=show_labels)
        out.write(annotated)

    cap.release()
    out.release()
    return out_path

# ---------------- Interface Streamlit ----------------
st.title("🧠 Stroke-IA – Détection AVC par IA")

# Sidebar (paramètres utilisateur)
st.sidebar.header("⚙️ Paramètres")
conf_threshold = st.sidebar.slider("Seuil de confiance", 0.1, 1.0, 0.25, 0.05)
show_labels = st.sidebar.checkbox("Afficher les labels", value=True)

st.sidebar.header("📂 Exemples rapides")
if st.sidebar.button("Tester une image exemple"):
    if os.path.exists("example.jpg"):
        img = Image.open("example.jpg")
        path = predict_image(img, conf=conf_threshold, show_labels=show_labels)
        st.image(path, caption="Exemple annoté", use_container_width=True)
    else:
        st.warning("⚠️ Aucun fichier example.jpg trouvé dans le projet.")

if st.sidebar.button("Tester une vidéo exemple"):
    if os.path.exists("example.mp4"):
        path = predict_video("example.mp4", conf=conf_threshold, show_labels=show_labels)
        st.video(path)
    else:
        st.warning("⚠️ Aucun fichier example.mp4 trouvé dans le projet.")

# Section vidéo upload
st.header("🎥 Détection sur vidéo")
video_file = st.file_uploader("Uploader une vidéo (mp4, mov, etc.)", type=["mp4", "mov"])
if video_file and st.button("Analyser la vidéo"):
    temp_path = os.path.join(SAVE_DIR, "temp_video.mp4")
    with open(temp_path, "wb") as f:
        f.write(video_file.read())
    result_path = predict_video(temp_path, conf=conf_threshold, show_labels=show_labels)
    st.video(result_path)

# Section image upload
st.header("🖼️ Détection sur image")
image_file = st.file_uploader("Uploader une image", type=["jpg", "jpeg", "png"])
if image_file and st.button("Analyser l'image"):
    image = Image.open(image_file)
    result_path = predict_image(image, conf=conf_threshold, show_labels=show_labels)
    st.image(result_path, caption="Image annotée", use_container_width=True)

# Disclaimer
st.markdown(f"""
---
⚠️ **Disclaimer :** Stroke-IA est une démo technique, pas un avis médical.  
© {datetime.now().year} — Badsi Djilali.
""")