Frequency-Domain Acoustic Features for Eggshell Crack Detection, 90 Eggs

A dataset of acoustic frequency-domain features generated from non-destructive resonance testing of 90 commercial chicken eggs, including 45 intact and 45 cracked eggs. The data was created by Abbas Rohani and includes features like Peak Frequency and Spectral Centroid extracted from signals transformed via FFT. It was last updated on 2026-05-25.

Use Cases

• Binary classification of eggshell integrity based on extracted spectral descriptors.
• Feature engineering and selection analysis for acoustic resonance signals.
• Reproducible research in machine learning-based non-destructive testing.
• Benchmarking acoustic sensing models for agricultural product quality.

Strengths

• Contains data from 90 eggs with a balanced class distribution (45 intact, 45 cracked).
• Features are extracted from a defined frequency range (500–20,000 Hz) using FFT.
• Includes specific spectral descriptors like Spectral Skewness and Spectral Kurtosis.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is small (216.0 KB), indicating a limited scope.

Provenance

Source

figshare

Collection Method

Generated from non-destructive acoustic resonance testing under controlled impact excitation.

Freshness

Last updated 2026-05-25 12:45:50; freshness should be verified.