import cv2
import numpy as np
import joblib
from skimage.feature import hog
import imutils
from imutils.object_detection import non_max_suppression
from skimage.transform import pyramid_gaussian


def sliding_window(image, stepSize, windowSize):
    for y in range(0, image.shape[0] - windowSize[1] + 1, stepSize):
        for x in range(0, image.shape[1] - windowSize[0] + 1, stepSize):
            yield (x, y, image[y:y + windowSize[1], x:x + windowSize[0]])


# HOG parameters (should match those used during training)
hog_params = {
    'orientations': 9,
    'pixels_per_cell': (8, 8),
    'cells_per_block': (2, 2),
    'block_norm': 'L2-Hys'
}

# Load the pre-trained model (grid search object containing the best estimator)
model = joblib.load('svm_natural_images_modelUTC.pkl')
# If the saved object is a GridSearchCV, extract the best estimator.
if hasattr(model, 'best_estimator_'):
    model = model.best_estimator_

# Detection parameters
windowSize = (64, 64)
hog_size = (64, 64)
stepSize = 8  # Step size for sliding window.
scale_factor = 1.01 # Scaling factor for image pyramid.
detection_threshold = 0.85  # Minimum probability to consider a detection valid.

# Load the test image (change the filename as needed)
image = cv2.imread("test3.jpg")

orig = image.copy()
detections = []

# Loop over the image pyramid
for resized in tuple(pyramid_gaussian(image, downscale=scale_factor, channel_axis=-1)):
    # The ratio of the current resized image relative to the original image
    scale_ratio = image.shape[1] / float(resized.shape[1])

    # Slide a window across the resized image
    for (x, y, window) in sliding_window(resized, stepSize, windowSize):
        # Ensure the window is of the desired size
        if window.shape[0] != windowSize[1] or window.shape[1] != windowSize[0]:
            continue

        if window.dtype != np.uint8:
            window = (window * 255).astype(np.uint8)


        gray = cv2.cvtColor(window, cv2.COLOR_BGR2GRAY)
        gray = cv2.resize(gray, hog_size, interpolation=cv2.INTER_AREA)

        features = hog(gray, **hog_params)
        features = features.reshape(1, -1)

        # Predict the class and probability for this window
        pred = model.predict(features)
        prob = model.predict_proba(features)[0]

        if 'person' in model.classes_:
            person_index = list(model.classes_).index('person')
            person_prob = prob[person_index]
        else:
            person_prob = 0

        if pred[0] == 'person' and person_prob > detection_threshold:
            startX = int(x * scale_ratio)
            startY = int(y * scale_ratio)
            endX = int((x + windowSize[0]) * scale_ratio)
            endY = int((y + windowSize[1]) * scale_ratio)
            detections.append((startX, startY, endX, endY))

if len(detections) > 0:
    detections = np.array(detections)
    pick = non_max_suppression(detections, probs=None, overlapThresh=0.3)
    # Draw the final bounding boxes on the original image.
    for (xA, yA, xB, yB) in pick:
        cv2.rectangle(orig, (xA, yA), (xB, yB), (0, 255, 0), 2)

# Display the detections
cv2.imshow("Detections", orig)
cv2.waitKey(0)
cv2.destroyAllWindows()