Abstract:Cultivar variability influences the near-infrared modeling and analysis of the internal quality of the fruit due to the different cell structure, composition and optical transmission characteristics of the fruit so that the original model can not predict the fruit quality parameters with high precision. The feasibility of a multi-cultivar model's development for the online determination of the internal quality of apple including "Candy Heart", "Red Fuji" and "Crystal Fuji" was investigated. Near infrared diffuse transmittance spectra of each cultivar were collected by the fruit sorting equipment under the condition of the interval time of 100 ms and motion speed of 5 s(-1). The spectral curves of all the cultivars were similar where the prominent absorption peaks were near 650, 709 and 810 nm, and troughs were near 670, 750 and 830 nm, and their variations were mainly reflected in the spectral absorption intensity. The spectral pre-process methods including multiplicative scatter correction, Savitzky-Golay smoothing and normalization were employed to filter out the variations in signals caused by the cultivars. Partial least squares regression method was used to establish the common model for the soluble solid content where the calibration sets of the total samples were combined. Uninformative variable elimination was used to select the modeling variables whose number of effective variables selected was 155, and the performance of the UVE-PLS model resulted in greater coefficient of determination for prediction of 0. 80, lower root mean square error of 0. 61% and higher residual prediction deviation of 2. 21. Successive projections algorithm was employed to select the variables in the wavelengths selected by UVE and the number of variables selected was 22. Multivariable linear regression was used to establish the simplified model, which resulted in coefficient of determination for prediction of 0. 78 and root mean square error of prediction of 0. 64%. The test sets of all the cultivars were used to access the performance of best universal model, which resulted in latent variables of 6 similar to 10, coefficient of determination for prediction of 0. 77 similar to 0. 79 and root mean square error of 0. 45%similar to 0. 75%. The results highlighted the potential of dynamic on-line sorting instruments for the testing of internal qualities of apples. The prediction range of the single cultivar model was expanded, and the robustness of prediction model among different cultivars were improved by establishing the common model. Appropriate variable selection methods can decrease the number of model variables, reduce the complexity of the model and ultimately increase the model rate. The development of the universal model of different cultivars for predicting internal quality has a good potential application in wavelength-limited near infrared spectroscopy equipment.

Modification of the Effect of Maturity Variation on Nondestructive Detection of Apple Quality Based on the Compensation Model.

Enhancing Fruit SSC Detection Accuracy Via a Light Attenuation Theory-Based Correction Method to Mitigate Measurement Orientation Variability

Enhancing model robustness through different optimization methods and 1-D CNN to eliminate the variations in size and detection position for apple SSC determination

Improving accuracy of prediction model for soluble solids content of watermelon by variable selection based on near-infrared spectroscopy

Prediction of Soluble Solids Content During Storage of Apples with Different Maturity Based on VIS / NIR Spectroscopy

Soluble solid predictive models of apples with different maturity levels

Development of Multi-Cultivar Models for Predicting the Soluble Solid Content and Firmness of European Pear (pyrus Communis L.) Using Portable Vis–nir Spectroscopy

Dynamic Nondestructive Detection Models of Apple Quality in Critical Harvest Period Based on Near-Infrared Spectroscopy and Intelligent Algorithms

Non-destructive Detection of Apple Maturity by Constructing Spectral Index Based on Reflectance Spectrum

Nondestructive Determination of Soluble Solids Content, Firmness, and Moisture Content of “longxiang” Pears During Maturation Using Near‐infrared Spectroscopy

Determination of bagged 'Fuji' apple maturity by visible and near-infrared spectroscopy combined with a machine learning algorithm

Non-destructive Prediction of Soluble Solids and Dry Matter Contents in Eight Apple Cultivars Using Near-Infrared Spectroscopy

Effect of Measurement Position on Prediction of Apple Soluble Solids Content (SSC) by an On-Line Near-Infrared (NIR) System

Internal Quality Detection of Apples during Late Developmental Period Based on Near-infrared Spectral Technology

An uncertainty sampling strategy based model updating method for soluble solid content and firmness prediction of apples from different years

RAPID DETERMINATION OF MATURITY IN APPLE USING OUTLIER DETECTION AND CALIBRATION MODEL OPTIMIZATION

Influences of Detection Position and Double Detection Regions on Determining Soluble Solids Content (SSC) for Apples Using On-line Visible/Near-Infrared (Vis/NIR) Spectroscopy

Development of Multi-Cultivar Universal Model for Soluble Solid Content of Apple Online Using Near Infrared Spectroscopy

On-line detection of soluble solids content of apples from different origins by visible and near-infrared spectroscopy

Rapid Identification of Apple Maturity Based on Multispectral Sensor Combined with Spectral Shape Features

Establishment of a multi-position general model for evaluation of watercore and soluble solid content in ‘Fuji’ apples using on-line full-transmittance visible and near infrared spectroscopy