Inspection of paddy seed varietal purity using machine vision and multivariate analysis

• DOI: 10.1016/j.jafr.2021.100109
• Publication Date: 2021-01-20T16:51:24Z
• Authors: Nadia Ansari; Sharmin Sultana R...; Afroz Jahan; Muhammad Ashik-E-...; Anisur Rahman
• Keywords: RGB color model; Color analysis; Machine Vision

Seed varietal purity is vital to establish a uniform plant population. If the seeds are impure, it creates an unhealthy population that brings labor-intensive crop production. In this study, rapid inspection method was established classify paddy seed based on using machine vision technique with multivariate analysis methods. Three varieties of were taken, namely - BR 11, BRRI dhan 28 and 29. The individual image captured RGB camera white LED lighting conditions in laboratory. An processing algorithm developed for extracting 20 important features (seven color features, nine morphological four textural features) from 375 images. next step, significant difference extracted data among studied variance analysis. Also, principal component performed explore separability varieties. Accordingly, variety classification models combination selected feature partial least squares-discriminant (PLS-DA), Support vector machine-classification (SVM-C) K-Nearest Neighbors (KNN) algorithm. During model development, seen more compare features. accuracy 83.8%, 93.9%, 87.2% achieved combined color, morphological, PLS-DA, SVM-C, KNN model, respectively. Finally, stated SVM-C could be used variety.