utils.py

# Links directly in scanner.py
from color_detector import predict_color

# Return list of LAB values for 9 dots
def sample_face_colors(cv2, frame, grid_centers):
    face_colors = []
    for (x, y) in grid_centers:
        # Getting avg LAB for ROI around dots
        roi = frame[y-10:y+10, x-10:x+10]  # A 20x20 region around individual dot
        roi_lab = cv2.cvtColor(roi, cv2.COLOR_BGR2LAB)
        lab_avg = cv2.mean(roi_lab)[:3]  # Only L, A, B
        lab_avg_rounded = tuple(map(int, lab_avg))
        face_colors.append(lab_avg_rounded)
    return face_colors

# Function to check for similar face scanned by the center piece of that face
# Uses Euclidean Distance
def is_similar_color(np, c1, c2, threshold=12):
    diff = np.linalg.norm(np.array(c1) - np.array(c2))
    return diff < threshold

# Classify face colors
def classify_face_colors(rf_model, scanned_faces):
    return ''.join(''.join(predict_color(rf_model, lab) for lab in face) for face in scanned_faces)

# Save the data collected to dataset.csv one-by-one
def save_to_csv(os, csv, LAB_values, label, dataset_path="dataset.csv"):
    file_exists = os.path.isfile(dataset_path)
    with open(dataset_path, mode="a", newline="") as file:
        writer = csv.writer(file)
        if not file_exists:
            writer.writerow(["L", "A", "B", "label"])  # header
        for LAB in LAB_values:
            writer.writerow([*LAB, label])