Contactless Classification of Strawberry Using Hyperspectral Imaging

Abstract

Rapid non-contact estimation of fruit quality parameters is an essential factor for an efficient food processing pipeline. We propose a novel workflow for the contactless classification of strawberries based on their sugar content, using Hyperspectral Imaging (HSI) and One-Dimensional Convolutional Neural Network (1D - CNN). Sugar content is an important quality aspect of strawberries, hence classification based on sugar content gives more yield to the fruit producers. We used Visible and Near Infrared (VNIR) hyperspectral camera to acquire HSI data of 50 ripe strawberries and applied the proposed method to classify them. To verify the advantage of the proposed method, the results from 1DCNN are compared against other standard classification methods such as Spectral Angle Mapper (SAM), and Spectral Information Divergence (SID). The results show that the 1D-CNN outperformed other methods by achieving 96.6% classification accuracy.