Coffee Roasting Agtron Dataset

Data Capture Conditions

• Device Variability: Coffee bean images are taken using different mobile devices to account for varying camera qualities and resolutions.
• Lighting Reference: Alongside the coffee images, a photo of a blank white sheet is included to serve as a lighting reference. This allows for normalization of lighting conditions, making the dataset reliable for visual analysis.

Agtron Scale and Its Importance

The Agtron scale is the industry standard for measuring the degree of coffee roasting. As the roast becomes darker, the value on the scale decreases. This scale is crucial in both quality control and research, allowing coffee professionals to consistently classify and compare roast levels.

Applications

This dataset is useful for several applications, including:

• Roast Level Analysis: Training machine learning models to identify and classify coffee roast levels based on visual inputs.
• Quality Control: Automating the analysis of roast consistency in coffee production.
• Research & Development: Coffee researchers and companies can utilize the dataset to innovate new roasting techniques or improve product consistency.

Potential Use in AI and Machine Learning

The dataset serves as a foundation for AI development, specifically in image recognition. With the reference lighting condition and varied device capture, this dataset is ideal for developing models that can automate the classification of coffee roast levels based on the Agtron scale.