Dark Mode
Dry Bean Morphological Attribute Dataset
Food & Beverage Consumption
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This collection of data details the geometric properties extracted from seven distinct varieties of dry beans, developed specifically for computer vision research. The purpose of this resource is to facilitate the development of classification models capable of distinguishing between these registered varieties, which often possess similar physical attributes, thereby aiding in uniform seed classification. The underlying research utilized a high-resolution camera system to capture images of 13,611 bean grains. Following imaging, the grains underwent segmentation and feature extraction stages, yielding 16 measurable features covering dimensions and shape characteristics.
Columns
The dataset includes 17 attributes, primarily representing dimension and shape features derived from image analysis:
- Area (A): The area occupied by the bean zone, measured by the count of pixels within its bounds.
- Perimeter (P): Defined as the length surrounding the bean circumference.
- Major axis length (L): The measurement between the endpoints of the longest line segment that can be drawn within the bean.
- Minor axis length (l): The longest line segment drawn within the bean while situated perpendicular to the major axis.
- Aspect ratio (K): Describes the proportional relationship between the Major axis length (L) and the Minor axis length (l).
- Eccentricity (Ec): The calculated eccentricity of an ellipse having identical moments to the region of the bean.
- Convex area (C): The total number of pixels contained within the smallest convex polygon that is able to enclose the bean seed area.
- Equivalent diameter (Ed): The calculated diameter of a perfect circle that shares the exact area of the bean seed area.
- Extent (Ex): The ratio of the pixels within the bounding box relative to the total area of the bean.
- Solidity (S): Also known as convexity, defined as the ratio comparing the pixels in the convex shell to the pixels found within the beans.
- Roundness (R): A metric calculated using the formula (4piA)/(P^2).
- Compactness (CO): Measures the roundness of the object, derived as Ed/L.
- ShapeFactor1 (SF1), ShapeFactor2 (SF2), ShapeFactor3 (SF3), ShapeFactor4 (SF4): Four derived shape factor variables.
- Class: The categorical target variable, listing the seven bean types: Seker, Barbunya, Bombay, Cali, Dermosan, Horoz, or Sira.
Distribution
The information is available in a typically structured format, such as CSV, with a file size of 2.48 MB. It consists of 13,611 valid records derived from individual bean grains. The structural integrity is high, with sources indicating zero missing or mismatched values across the 17 attributes. The expected update frequency for this resource is 'Never', indicating a static compilation.
Usage
This data is ideally suited for Machine Learning applications, particularly those focused on multi-class classification problems. It is a robust resource for developing and evaluating algorithms in agricultural sorting, quality control, and automated classification. Specific applications include Exploratory Data Analysis, Clustering algorithms, and building supervised models to classify objects based on geometric and physical characteristics.
Coverage
The scope of this data is focused on the morphometric characteristics of seven registered varieties of dry beans: Seker, Barbunya, Bombay, Cali, Dermosan, Horoz, and Sira. No specific geographical location or temporal range is noted for the data acquisition, as the focus is purely on the distinguishing features of the bean types themselves.
License
CC0: Public Domain
Who Can Use It
The primary audience includes:
- Data Scientists: For training and testing classification and clustering algorithms.
- Machine Learning Engineers: Those specializing in computer vision for feature extraction and pattern recognition tasks.
- Agricultural Technologists: Professionals seeking to implement automated sorting and grading systems for crop quality assessment.
- Students and Academics: Utilizing the data as a benchmark for research in pattern recognition and machine learning studies.
Dataset Name Suggestions
- Dry Bean Computer Vision Classification Data
- Dry Bean Morphological Attribute Dataset
- Agricultural Seed Feature Data
- Seven Bean Varieties Classification Set
Attributes
Original Data Source: Dry Bean Morphological Attribute Dataset
QUALITYLoading...