Whiskers Segmentation Dataset

Project Overview:

Objective

To create a dataset focused on the segmentation of whiskers on various animals, aiming to support research in animal behavior, morphological studies, and species identification applications

Scope

Collect images featuring animals known for prominent whiskers, such as cats, seals, rats, and more. Annotations will segment each whisker, and provide metadata regarding its length, curvature, and orientation

Cat Dataset
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset

Sources

  • Engaged with wildlife photographers and enthusiasts to gather a carefully collected and successfully curated selection of images.
  • Collaborated with veterinary clinics and animal rehabilitation centers, resulting in a meticulously collected and thoughtfully curated assortment of visuals.
  • Utilized zoo and aquarium archives, contributing to a successfully collected and historically rich repository of wildlife images.
  • Established connections with research institutes focused on mammalian studies, leading to a carefully collected and scientifically relevant dataset.
Cat Dataset
Whiskers Segmentation Dataset

Data Collection Metrics

  • Total Whisker Images: 280,000
  • Domestic Cats: 100,000
  • Rats and Mice: 50,000
  • Seals and Sea Lions: 60,000
  • Other Mammals: 70,000

Annotation Process

Stages

  1. Image Pre-processing: Enhancement for better clarity and contrast.
  2. Whisker Segmentation: Precisely outlining the boundary of each whisker using segmentation masks.
  3. Metadata Annotation: Recording details such as whisker length, curvature angle, base thickness, and observed functionality (e.g., tactile, directional).
  4. Validation: Cross-checking annotations using a blend of expert opinions and initial whisker detection models.

Annotation Metrics

  • Total Whisker Segmentation Annotations: 1,120,000 (assuming an average of 4 prominent whiskers per image)
  • Metadata Annotations: 1,120,000
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset

Quality Assurance

Stages

Automated Whisker Recognition Verification: Using early models to check segmented whiskers for consistency.
Peer Review: Double-checking annotations through a secondary set of annotators.
Inter-annotator Agreement: Randomly selecting images for re-annotation by multiple individuals to guarantee annotation uniformity.

QA Metrics

  • Annotations Validated using Whisker Recognition: 140,000 (50% of total images)
  • Peer Reviewed Annotations: 84,000 (30% of total images)
  • Inconsistencies Identified and Corrected: 5,600 (2% of total images)

Conclusion

The Whiskers Segmentation Dataset is an invaluable contribution to the domain of animal research. With its emphasis on detail and diversity, it promises to facilitate groundbreaking studies in animal behavior, physiology, and evolutionary biology. By leveraging this dataset, researchers and scientists can gain more profound insights into the nuanced world of whisker functionalities across various mammals.

quality dataset

Quality Data Creation

Guaranteed TAT‚Äč

Guaranteed TAT

ISO 9001:2015, ISO/IEC 27001:2013 Certified‚Äč

ISO 9001:2015, ISO/IEC 27001:2013 Certified

HIPAA Compliance‚Äč

HIPAA Compliance

GDPR Compliance‚Äč

GDPR Compliance

Compliance and Security‚Äč

Compliance and Security

Let's Discuss your Data collection Requirement With Us

To get a detailed estimation of requirements please reach us.

Scroll to Top