Pommes Pont-Neuf

Dataset Description: Good and Bad French Fries Classification

1. Overview

This dataset comprises over 500 samples of French fries (pommes pont-neuf) categorized as good or bad based on their visual and qualitative attributes. The dataset is designed for machine learning applications, particularly for training and evaluating models in image-based food quality assessment.

2. Data Collection

Device Used: Redmi 10 Prime mobile camera

Lighting Condition: Natural daylight

Background: White

Number of Samples: More than 500

Categories:

Good French Fries: Fries with uniform shape, golden-brown color, crisp texture, and no visible burns or defects.

Bad French Fries: Fries that are overcooked (burnt edges), undercooked (pale appearance), irregularly shaped, broken, or with visible black spots.

3. Image Specifications

Resolution: Based on the Redmi 10 Prime's camera specifications, the images are expected to have high clarity, making it easier to distinguish quality differences.

Format: Likely captured in JPEG or PNG format.

Framing: Single fry or multiple fries per image, uniformly arranged against a white background.

4. Applications

Automated food quality classification

Training deep learning models for food inspection

Enhancing fast-food industry quality control

Computer vision applications in food safety

This dataset provides a solid foundation for AI-driven quality assessment of French fries, ensuring consistency and efficiency in food evaluation.