SignLanguageRecognition/dataLabeling.py at main · baranzeyn/SignLanguageRecognition · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
import os
import csv
import cv2 as cv
import mediapipe as mp

# Veri seti yolu ve CSV kaydetme yolu
data_dir = '/Users/zeynepbaran/Desktop/Sign Language Recognition/data'
output_csv = '/Users/zeynepbaran/Desktop/Sign Language Recognition/coordinates.csv'

# Mediapipe Hands hazırlığı
mp_hands = mp.solutions.hands
hands = mp_hands.Hands(static_image_mode=True, max_num_hands=2, min_detection_confidence=0.5)

# CSV dosyasını başlat
with open(output_csv, mode='w', newline='', encoding='utf-8') as f:
    writer = csv.writer(f)
    # Başlıklar: Sağ ve sol el için x, y koordinatları + etiket
    headers = (
        [f'right_x{i}' for i in range(21)] + [f'right_y{i}' for i in range(21)] +
        [f'left_x{i}' for i in range(21)] + [f'left_y{i}' for i in range(21)] +
        ['label']
    )
    writer.writerow(headers)

    # Her kategorideki resimleri işleyelim
    for label in os.listdir(data_dir):
        label_path = os.path.join(data_dir, label)
        if not os.path.isdir(label_path):
            continue

        print(f"Processing label: {label}")

        for image_name in os.listdir(label_path):
            image_path = os.path.join(label_path, image_name)
            if not image_name.lower().endswith(('png', 'jpg', 'jpeg')):
                continue

            # Resmi yükle
            image = cv.imread(image_path)
            if image is None:
                print(f"Skipping invalid image: {image_path}")
                continue

            # Resmi BGR'den RGB'ye çevir
            rgb_image = cv.cvtColor(image, cv.COLOR_BGR2RGB)

            # Mediapipe ile işleme
            results = hands.process(rgb_image)

            # Sağ ve sol el için varsayılan koordinat listeleri (sıfırlarla dolu)
            right_hand_coords = [0] * 42  # 21 x (x, y)
            left_hand_coords = [0] * 42

            if results.multi_hand_landmarks:
                for hand_landmarks, handedness in zip(results.multi_hand_landmarks, results.multi_handedness):
                    # Sağ ve sol eli ayır
                    if handedness.classification[0].label == 'Right':
                        right_hand_coords = []
                        reference_point = hand_landmarks.landmark[0]
                        for lm in hand_landmarks.landmark:
                            right_hand_coords.append(lm.x - reference_point.x)
                            right_hand_coords.append(lm.y - reference_point.y)
                    elif handedness.classification[0].label == 'Left':
                        left_hand_coords = []
                        reference_point = hand_landmarks.landmark[0]
                        for lm in hand_landmarks.landmark:
                            left_hand_coords.append(lm.x - reference_point.x)
                            left_hand_coords.append(lm.y - reference_point.y)

            # Koordinatları ve etiketi yaz
            writer.writerow(right_hand_coords + left_hand_coords + [label])

print(f"Relative coordinates saved to {output_csv}")