File size: 5,064 Bytes
9eeef69 f9ed0a9 |
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 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 |
---
dataset_info:
features:
- name: image_id
dtype: int64
- name: image
dtype: image
- name: width
dtype: int32
- name: height
dtype: int32
- name: objects
struct:
- name: bbox
sequence:
sequence: float32
length: 4
- name: category_id
sequence: int64
- name: category
sequence: string
- name: area
sequence: float32
- name: iscrowd
sequence: int64
splits:
- name: train
num_bytes: 97076536.768
num_examples: 3456
- name: valid
num_bytes: 14929397.0
num_examples: 470
- name: test
num_bytes: 9853454.0
num_examples: 311
download_size: 121355015
dataset_size: 121859387.768
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*
---
# Experimento-3 - Industrial Machinery Text Detection Dataset
## Dataset Description
This dataset contains **4,237 images** of industrial machinery nameplates with detailed text field annotations for OCR and information extraction tasks. The dataset focuses on extracting key information from equipment nameplates including manufacturer, model, serial numbers, and dates.
## Dataset Summary
- **Task**: Industrial text detection and OCR
- **Domain**: Industrial machinery and equipment
- **Images**: 4,237 total images
- **Annotations**: Bounding boxes for text fields with 8 categories
- **Source**: [Roboflow Universe - Experimento-3](https://universe.roboflow.com/marcos-feria/experimento-3)
## Dataset Structure
### Splits
| Split | Images |
|-------|--------|
| Train | 3,456 |
| Valid | 470 |
| Test | 311 |
### Categories
The dataset includes 8 text field categories commonly found on industrial equipment nameplates:
| ID | Category | Description |
|----|----------|-------------|
| 0 | tipos-pl | Equipment type (unused in annotations) |
| 1 | FABRICANTE | Manufacturer name |
| 2 | FECHA | Date information |
| 3 | MODEL | Model designation |
| 4 | MODELO | Model designation (Spanish) |
| 5 | NUMERO DE SERIE | Serial number (full text) |
| 6 | SN | Serial number (abbreviated) |
| 7 | YEAR | Year information |
## Data Fields
Each example contains:
- `image_id`: Unique image identifier
- `image`: PIL Image of the machinery nameplate
- `width`: Image width in pixels
- `height`: Image height in pixels
- `objects`: Dictionary containing:
- `bbox`: List of bounding boxes in [x, y, width, height] format (COCO format)
- `category_id`: List of category IDs (0-7)
- `category`: List of category names
- `area`: List of bounding box areas
- `iscrowd`: List of crowd flags (typically 0)
## Use Cases
This dataset is ideal for:
1. **Industrial OCR Systems**: Extracting text from machinery nameplates
2. **Equipment Inventory Management**: Automated data collection from equipment
3. **Maintenance Planning**: Identifying equipment details for service schedules
4. **Asset Tracking**: Digital cataloging of industrial equipment
5. **Computer Vision Research**: Multi-language text detection in industrial settings
## Data Collection
The images were collected from various industrial machinery and equipment, focusing on nameplates and identification tags. The dataset includes equipment from multiple manufacturers and spans different time periods, providing diverse examples for robust model training.
## Licensing & Attribution
Please refer to the original Roboflow dataset for licensing information. When using this dataset, please cite:
```bibtex
@misc{
experimento-3_dataset,
title = { Experimento-3 Dataset },
type = { Open Source Dataset },
author = { Marcos Feria },
howpublished = { \url{ https://universe.roboflow.com/marcos-feria/experimento-3 } },
url = { https://universe.roboflow.com/marcos-feria/experimento-3 },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2023 },
note = { visited on 2025-01-25 },
}
```
## Loading the Dataset
```python
from datasets import load_dataset
# Load the full dataset
dataset = load_dataset("kahua-ml/experimento3-industrial-text-detection")
# Load specific split
train_dataset = load_dataset("kahua-ml/experimento3-industrial-text-detection", split="train")
# Example usage
example = dataset["train"][0]
image = example["image"]
bboxes = example["objects"]["bbox"]
categories = example["objects"]["category"]
```
## Dataset Statistics
- **Average annotations per image**: ~6.2 text fields
- **Most common categories**: FABRICANTE, SN, NUMERO DE SERIE
- **Image resolution**: Primarily 544x416 pixels
- **Languages**: Mixed Spanish/English text fields
## Applications
This dataset has been successfully used for:
- Training YOLO models for industrial text detection
- Fine-tuning vision transformers for equipment OCR
- Developing maintenance automation systems
- Creating inventory management solutions
## Contact
For questions about this dataset, please refer to the original Roboflow project or create an issue in this repository.
|