PCB-Defects-Balanced-3Class Computer Vision Dataset
About PCB-Defects-Balanced-3Class Dataset
Balanced three‑class PCB defect dataset for classification and evaluation.
- Classes: missing_hole, mouse_bite, short
- Images per class: 100 source images each (balanced distribution)
- Split per class: 80 training, 15 validation, 5 test
- Preprocessing: No augmentation, resizing, or contrast adjustments applied Source: PCB Defects Dataset (Roboflow Universe, Public Domain)
Class Definitions Missing Hole: Drill holes required for component placement or vias are absent. This defect disrupts connectivity and prevents proper assembly, often caused by errors in drilling or incomplete fabrication.
Mouse Bite:
Small, irregular notches or semi‑circular gaps along the edges of the PCB, resembling bite marks. Typically caused by breakout tabs during panel separation, they can weaken board edges and affect mechanical stability.
Short:
Unintended electrical connection between two or more conductive traces or pads. Shorts are usually caused by solder bridging, excess copper, or fabrication errors, and can lead to circuit malfunction or failure.
Use Free Short, Missing_hole and Mouse_bite Detection API
Powered by general detection model
pip install inference-sdk# 1. Import the library
from inference_sdk import InferenceHTTPClient
# 2. Connect to your workspace
client = InferenceHTTPClient(
api_url="https://serverless.roboflow.com",
api_key="API_KEY"
)
# 3. Run your workflow on an image
result = client.run_workflow(
workspace_name="<YOUR_WORKSPACE>",
workflow_id="<YOUR_WORKFLOW_ID>",
images={
"image": "YOUR_IMAGE.jpg" # Path to your image file
},
parameters={
"classes": "short, missing_hole, mouse_bite"
},
use_cache=True # cache workflow definition for 15 minutes
)
# 4. Get your results
print(result)Run on custom image
Drop an image here or click to upload
Tell the agent what you want to build.
Cite This Project
LicensePublic DomainIf you use this dataset in a research paper, please cite it using the following BibTeX:
@misc{ pcb-defects-balanced-3class-yuc6z_dataset,
title = { PCB-Defects-Balanced-3Class Dataset },
type = { Open Source Dataset },
author = { sarawans Workspace },
howpublished = { \url{ https://universe.roboflow.com/sarawans-workspace/pcb-defects-balanced-3class-yuc6z } },
url = { https://universe.roboflow.com/sarawans-workspace/pcb-defects-balanced-3class-yuc6z },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2026 },
month = { jul },
note = { visited on 2026-07-29 },
}