Datasets:

menasi11
/

colorbar-range-dataset

license: mit
task_categories:
  - image-text-to-text
  - image-feature-extraction
language:
  - en
pretty_name: Colorbar Range Dataset
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: image_id
      dtype: string
    - name: label
      struct:
        - name: Blue
          sequence: int64
        - name: Green
          sequence: int64
        - name: Yellow
          sequence: int64
        - name: Orange
          sequence: int64
        - name: Dark Orange
          sequence: int64
        - name: Red
          sequence: int64
        - name: Dark Red
          sequence: int64
        - name: Magenta
          sequence: int64
        - name: Pink
          sequence: int64
        - name: Purple
          sequence: int64
    - name: image
      dtype: image
  splits:
    - name: train
      num_bytes: 956621
      num_examples: 80
  download_size: 942355
  dataset_size: 956621

Colorbar Range Dataset

This dataset contains vertical colorbar images annotated with structured numeric ranges for each color segment. These colorbars are commonly used in scientific plots, heatmaps, and machine learning visualizations.

Each image is paired with metadata describing the numeric range represented by each color.

Dataset Structure

Each sample contains:

image : Colorbar image (PNG)
image_id : Unique identifier for the image
label (or ranges) : Dictionary mapping color name → numeric range

Example Annotation

{
  "Blue": [0, 10],
  "Green": [10, 15],
  "Yellow": [15, 20],
  "Orange": [20, 25],
  "Red": [25, 30]
}

license: mit task_categories: - image-text-to-text - image-feature-extraction language: - en pretty_name: Colorbar Range Dataset

Colorbar Range Dataset

How to Use

1. Install Dependencies

pip install datasets pillow

2. Load the Dataset from Hugging Face

from datasets import load_dataset

dataset = load_dataset(
    "menasi11/colorbar-range-dataset",
    split="train"
)

sample = dataset[0]
sample["image"].show()