from flask import Flask, request, jsonify, make_response
import numpy as np
import cv2
from skimage.feature import local_binary_pattern
import os
from joblib import load

app = Flask(__name__)


def extract_lbp_features(image, radius=1, n_points=8, method='uniform'):
    lbp = local_binary_pattern(image, n_points, radius, method)
    (hist, _) = np.histogram(lbp.ravel(), bins=np.arange(0, n_points + 3), range=(0, n_points + 2))
    # Normalize the histogram
    hist = hist.astype("float")
    hist /= (hist.sum() + 1e-7)
    return hist


# Mendapatkan path lengkap ke file facemodel.joblib
model_path = os.path.join(os.path.dirname(__file__), 'facemodel.joblib')
# Memuat model dari file
knn_model = load(model_path)


@app.route('/classify', methods=['POST'])
def classify_face():
    # Mendapatkan gambar dari request
    file = request.files['image']

    # Membaca gambar
    image_bytes = file.read()
    original_image = cv2.imdecode(np.frombuffer(image_bytes, np.uint8), cv2.IMREAD_COLOR)

    gray_image = cv2.cvtColor(original_image, cv2.COLOR_BGR2GRAY)
    gray_image = cv2.equalizeHist(gray_image)

    face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
    faces = face_cascade.detectMultiScale(gray_image, scaleFactor=1.2, minNeighbors=10, minSize=(50, 50))

    predictions = []
    lbp_features = []

    for i, (x, y, w, h) in enumerate(faces, 1):
        cropped_face = original_image[y:y + h, x:x + w]
        resized_face = cv2.resize(cropped_face, (100, 100))
        gray_resized_face = cv2.cvtColor(resized_face, cv2.COLOR_BGR2GRAY)
        normalized_face = cv2.normalize(gray_resized_face, None, alpha=0, beta=255, norm_type=cv2.NORM_MINMAX)
        test_lbp_features = extract_lbp_features(normalized_face, radius=1, n_points=8, method='uniform')
        lbp_features.append(test_lbp_features.tolist())
        predicted_class = knn_model.predict([test_lbp_features])[0]
        predictions.append(predicted_class)  # Hanya menambahkan kelas prediksi

    response = make_response(jsonify(predictions))
    response.headers['X-LBP-Features'] = str(lbp_features)
    return response


if __name__ == '__main__':
    app.run(debug=True)