Whiskers Segmentation Dataset

Project Overview:

Objective

To create a dataset focused on the analysis of side burn on various animals, aiming to support research in animal behavior, definition studies, and species identification applications

Scope

 Annotations will segment each side burn, and provide tag regarding its length,bend, and orientation

Whiskers Segmentation Dataset
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset
Whiskers Segmentation Dataset

Sources

Collaborated with pet clinics and animal gret back centers, resulting in a exactly collected and intentionally curated array of visuals.

  • Utilized zoo and fish bowl archives, related 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.
case study-post
Whiskers Segmentation 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. 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.

Technology

Quality Data Creation

Technology

Guaranteed TAT

Technology

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

Technology

HIPAA Compliance

Technology

GDPR Compliance

Technology

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