نظرة عامة على مجموعات بيانات التجزئة الدلالية#

Semantic segmentation assigns one class label to every pixel in an image. Unlike instance segmentation, semantic segmentation does not separate individual objects of the same class. The training target is a dense class map where each pixel stores a class ID.

يشرح هذا الدليل تنسيق مجموعة البيانات المستخدم بواسطة نماذج التجزئة الدلالية من Ultralytics YOLO ويسرد تهيئات مجموعات البيانات المضمنة المتاحة للتدريب والتحقق.

تنسيقات مجموعات البيانات المدعومة#

Two label formats are supported. The dataset loader picks PNG masks when the dataset YAML defines a masks_dir key, or when a masks/ folder already exists next to your images at the dataset root; otherwise it falls back to YOLO polygon labels.

تنسيق قناع PNG#

تستخدم مجموعات بيانات التجزئة الدلالية ملف صورة واحداً وملف قناع واحداً لكل عينة. القناع عبارة عن صورة ذات قناة واحدة، عادة ما تكون بصيغة PNG، حيث تمثل كل قيمة بكسل فهرس الفئة لبكسل الصورة المقابل.

• Pixel values 0, 1, 2, ... represent class IDs from the dataset names mapping.
• Pixel value 255 is treated as the ignore label and is excluded from loss and metric computation.
• Mask files should use the same stem as their matching image file, for example frankfurt_000000_000294.png.
• Masks are resolved as .png by default; if missing, other supported image extensions are also accepted. Use lossless formats such as .png or .tiff, since lossy compression (e.g. .jpg) corrupts the class ID pixel values.

The default layout keeps images and masks in parallel folders. The masks_dir value from the dataset YAML replaces the images path component to find masks.

dataset/
├── images/
│   ├── train/
│   └── val/
└── masks/
    ├── train/
    └── val/

For example, an image at images/train/aachen_000000_000019.png is paired with a mask at masks/train/aachen_000000_000019.png when masks_dir: masks.

تنسيق تسمية المضلعات لـ YOLO#

If your dataset already has Ultralytics YOLO polygon labels (one .txt per image with <class-index> <x1> <y1> <x2> <y2> ... rows), you can train semantic segmentation directly from them — no PNG mask conversion needed. See the instance segmentation dataset format for the row-level layout.

This path is selected automatically when the dataset YAML omits masks_dir and no masks/ folder exists next to your images at the dataset root — remove or rename any leftover masks/ folder, or the loader falls back to PNG-mask mode and looks for masks there instead. Behavior:

• يتم تحويل المضلعات إلى قناع دلالي لكل صورة وقت التحميل، مرتبة حسب المساحة بحيث تتجاوز الأجسام الأصغر الأجسام الأكبر في مناطق التداخل.
• Multi-class (N > 1 in names): an extra background class is appended after your declared classes for pixels not covered by any polygon. The model is built with N + 1 output channels and the last channel is background.
• Single-class (N == 1 in names): still trained as 1 class. The mask is binary, with your declared class shown as 1 and pixels not covered by any polygon as 0. No extra background class is added to names.
• Pixels added by augmentation padding (e.g. random crop) still use 255 as the ignore label.

استخدم هذا المسار عندما تكون بياناتك مصنفة بالفعل كمضلعات مثيلات وتريد نموذج تجزئة دلالية من نفس الملفات.

تنسيق YAML لمجموعة البيانات#

يتم تهيئة مجموعات بيانات التجزئة الدلالية باستخدام ملفات YAML. الحقول الرئيسية هي:

# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license

# Cityscapes semantic segmentation dataset (19 classes)
# Documentation: https://docs.ultralytics.com/datasets/semantic/cityscapes8
# Example usage: yolo semantic train data=cityscapes8.yaml model=yolo26n-sem.pt
# parent
# ├── ultralytics
# └── datasets
#     └── cityscapes8 ← downloads here (small subset)
#         └── images
#         └── masks

# Dataset root directory
path: cityscapes8 # dataset root dir
train: images/train # train images (relative to 'path') 4 images
val: images/val # val images (relative to 'path') 4 images

masks_dir: masks # semantic mask directory

# Cityscapes 19-class labels
names:
  0: road
  1: sidewalk
  2: building
  3: wall
  4: fence
  5: pole
  6: traffic light
  7: traffic sign
  8: vegetation
  9: terrain
  10: sky
  11: person
  12: rider
  13: car
  14: truck
  15: bus
  16: train
  17: motorcycle
  18: bicycle

# Map source label IDs to train IDs; ignore_label is converted to 255.
label_mapping:
  -1: ignore_label
  0: ignore_label
  1: ignore_label
  2: ignore_label
  3: ignore_label
  4: ignore_label
  5: ignore_label
  6: ignore_label
  7: 0
  8: 1
  9: ignore_label
  10: ignore_label
  11: 2
  12: 3
  13: 4
  14: ignore_label
  15: ignore_label
  16: ignore_label
  17: 5
  18: ignore_label
  19: 6
  20: 7
  21: 8
  22: 9
  23: 10
  24: 11
  25: 12
  26: 13
  27: 14
  28: 15
  29: ignore_label
  30: ignore_label
  31: 16
  32: 17
  33: 18

# Download URL (optional)
download: https://github.com/ultralytics/assets/releases/download/v0.0.0/cityscapes8.zip

Use label_mapping when the source mask IDs do not already match contiguous training class IDs. Cityscapes and ADE20K include mappings that convert original label IDs into YOLO semantic segmentation train IDs and ignore unused labels.

الاستخدام#

قم بتدريب نموذج تجزئة دلالية YOLO26 باستخدام Python أو CLI:

from ultralytics import YOLO

# Load a pretrained semantic segmentation model
model = YOLO("yolo26n-sem.pt")

# Train on the Cityscapes8 semantic segmentation dataset
results = model.train(data="cityscapes8.yaml", epochs=100, imgsz=1024)

مجموعات البيانات المدعومة#

Ultralytics provides semantic segmentation dataset YAML files for these datasets. See the semantic segmentation task page for the full pretrained-model benchmark table.

• Cityscapes: مجموعة بيانات تجزئة دلالية لمشاهد الشوارع الحضرية مع 19 فئة للتدريب.
• Cityscapes8: An 8-image Cityscapes subset for quick tests and CI checks.
• ADE20K: مجموعة بيانات تحليل المشاهد مع 150 فئة دلالية.

إضافة مجموعة البيانات الخاصة بك#

الخيار أ — أقنعة PNG#

1. Save your images under split folders such as images/train and images/val.
2. Save one single-channel mask per image under the mirrored mask folders, such as masks/train and masks/val.
3. Ensure mask pixel values are class IDs. Use 255 for pixels that should be ignored.
4. Create a dataset YAML with path, train, val, masks_dir, and names.
5. Add label_mapping only when your mask IDs need conversion to contiguous train IDs.

path: path/to/my-semantic-dataset
train: images/train
val: images/val
masks_dir: masks

names:
    0: background
    1: road
    2: building

الخيار ب — تسميات المضلعات#

1. Lay out images and .txt polygon files exactly as for instance segmentation.
2. Create a dataset YAML with path, train, val, and names — omit masks_dir.
3. Make sure no masks/ folder exists next to your images at the dataset root — its presence alone switches the loader to PNG-mask mode even without masks_dir in the YAML.
4. Do not add a "background" entry to names. For multi-class datasets the loader appends one automatically; for single-class datasets training stays at 1 class — your declared class becomes 1 in the mask and uncovered pixels become 0.

path: path/to/my-polygon-dataset
train: images/train
val: images/val

names:
    0: person
    1: car

Ultralytics Platform provides a polygon annotation tool for the semantic task, plus SAM-assisted Smart annotation — annotate directly in the browser and export or train on the resulting polygon-labeled dataset without setting up this layout by hand.

الأسئلة الشائعة#