The Use of Probabilistic Neural Network and UV Reflectance Spectroscopy as an Objective Cultured Pearl Quality Grading Method

Abstract

Pearl quality and value are determined as a combination of different features, with mollusk species, nacre thickness, luster, surface, shape, color and pearl size, being the most important. A pearl grader has to quantify visual observations and to assign a grading level to a pearl. The aim of this work was to reduce subjectivity in the assessment of some aspects of pearl quality by using artificial neural networks to predict pearl quality parameters from UV reflectance spectra. Given the good predictability of our previous model that used multilayer perceptron ANN modeling of UVVisible spectra to predict the grade of pearls, we wanted to simplify and improve the model by reducing the spectral input to UV only and by using classifier neural network modelling. It is hypothesized that as UV light is of higher energy than visible light, it may penetrate further into the surface of the pearl, and hence the corresponding UV diffuse reflectance spectrum may provide more information that can be used to assess pearl quality.The developed models were successful in predicting mollusk pearl growing species, pearl and donor color, luster, and surface complexity. The simplified models have been built resulting in more accurate prediction of selected pearl quality parameters when compared with the previous reported model.

Genomic Selection in Aquaculture: Application, Limitations and Opportunities With Special Reference to Marine Shrimp and Pearl Oysters

“…When applying this technology to non-invasive automated on-farm flow-through MVS, fish length has accurately been estimated in the rainbow trout (Miranda and Romero, 2017), as with fish mass in Jade Perch (Viazzi et al., 2015) with low relative mean errors (5.2 and 6.0%, respectively). Furthermore, fish skin or fillet color and pearl quality traits (e.g., color, luster, and complexion), which are traditionally recorded as categorical traits, can now be recorded as highly reliable continuous quantitative traits based on ultraviolet-vis spectrophotometry measurements (e.g., Urban et al., 2013; Kustrin and Morton, 2015). The performance of MVS and traditional color measurements has been compared with Atlantic salmon, rainbow trout, and pearls with spectral patterns produced by MSV more representative and consistent of the real color (Yagiz et al., 2009; Mamangkey et al., 2010; Colihueque et al., 2011, 2017; Toyota and Nakauchi, 2013), which will ultimately improve genomic selection predictions.…”

Application of second-generation sequencing (SGS) and third generation sequencing (TGS) in aquaculture breeding program

“…More recently, mechanized automation, imaging, and software developments have been utilized to rapidly acquire highresolution and high-volume quantitative phenotype data. Automated computer vision enables estimation of fish length in rainbow trout (Miranda and Romero, 2017) and fish mass in Jade Perch, Scortum barcoo (Viazzi et al, 2015) with very low residual errors, while UV-Vis spectrophotometry recorded fish skin colour and pearl quality traits (colour, lustre, completion) as continuous quantitative traits (Agatonovic-Kustrin and Morton, 2015). Monitoring and 2D or 3D imaging via machine vision systems (MVSs) has increased the selection accuracy (Saberioon et al, 2017;Zion, 2012).…”

Lombok Pearl Quality Classification Using a Combination of Feature Extraction and Artificial Neural Networks Based on Shape

“…This study reveals that their findings may serve as another way to understand research trends in mollusc pearl shells and contribute to future studies. (Agatonovic, 2015) using Probabilistic Neural Networks and UV Reflectance Spectroscopy as Pearl Quality Assessment Methods. From the research results, the developed model can predict the type of mollusc pearl.…”