Good and Bad classification of Egg Bread Toast
Description
Data Description for the Project: "Good and Bad Classification of Egg and Bread Toast" This dataset is designed to support a machine learning project aimed at classifying eggs and bread toasts as either "Good" or "Bad" based on various quality parameters. It comprises a total of 1000 samples, evenly distributed across two categories: Egg Samples: 500 total (250 Good, 250 Bad) Bread Toast Samples: 500 total (250 Good, 250 Bad) The dataset is structured to facilitate binary classification tasks and can be applied to food quality assessment systems, automated inspection lines, and educational projects related to food safety and processing. 1. Egg Samples Total Samples: 500 Good Eggs (250 samples): Fresh eggs with optimal physical and chemical properties. Bad Eggs (250 samples): Eggs that are spoiled, stale, or not suitable for consumption. Features: Shell Characteristics: Color: Ranges from white to brown shades. Texture: Smooth (good) vs. rough or cracked (bad). Cleanliness: Clean shell (good) vs. dirty/stained shell (bad). Internal Quality Parameters: Yolk Position & Shape: Centered and round yolk (good) vs. flattened or displaced yolk (bad). Albumen Consistency: Thick and firm albumen (good) vs. watery or thin albumen (bad). Odor: Neutral smell (good) vs. sulfuric or unpleasant odor (bad). Physical Tests: Float Test Results: Sinks and lies flat (good) vs. floats in water (bad). Weight: Standard weight range for fresh eggs vs. underweight or dehydrated eggs. 2. Bread Toast Samples Total Samples: 500 Good Bread Toast (250 samples): Properly toasted slices, evenly browned, and suitable for consumption. Bad Bread Toast (250 samples): Over-toasted (burnt), under-toasted (raw), or stale toasts. Features: Color and Texture: Color Uniformity: Golden-brown (good) vs. burnt black or pale (bad). Texture: Crisp outer layer with soft inner crumb (good) vs. hard, burnt, or soggy texture (bad). Moisture Content: Optimal moisture retained in good toasts vs. excessively dry or too moist in bad toasts. Physical Dimensions: Thickness: Uniform slices in good samples vs. uneven or broken slices in bad ones. Aroma and Taste: Pleasant toasted aroma and flavor (good) vs. burnt smell or off-flavor (bad). 3. Data Collection and Preprocessing All samples were collected under controlled conditions to ensure consistency. Visual inspection, olfactory tests, and physical assessments were performed. Images of each sample were captured under standardized lighting and angles. Data preprocessing involved normalization, feature extraction (color histograms, texture analysis), and removal of outliers. 4. Potential Applications Automated sorting systems in food industries.