Application Case | ATP2000H Spectrometer: High-Precision Prediction of Apple Firmness

In apple quality assessment, firmness serves as a critical indicator of texture and ripeness. Traditional detection methods relying on manual puncture tests are not only inefficient and destructive but also challenging to scale for standardized, high-volume testing. With the rapid advancement of spectroscopic technology, non-destructive, rapid, and high-precision apple firmness detection has now become a reality. In a recent study, researchers successfully achieved fast, non-destructive, and highly accurate prediction of apple firmness using the Optosky ATP2000H miniature fiber optic spectrometer.

01 Reliable Data Acquisition: ATP2000H Lays the Research Foundation

The study selected 1,022 apple samples from three varieties: Fuji, Gala, and Golden Delicious. Using the ATP2000H spectrometer, spectral data were collected from six marked points on each sample within the 400–700 nm wavelength range, yielding a total of 6,132 valid spectral datasets.

Throughout this process, the ATP2000H played a central role in core data acquisition. Its stable performance ensured the accuracy and consistency of the raw spectral data, providing a reliable foundation for subsequent modeling.

02 Efficient Modeling & Analysis: Algorithms Enable Accurate Prediction

After obtaining the spectral data collected by the ATP2000H, researchers performed data processing and modeling:

• Preprocessing: Applied Savitzky-Golay (SG) smoothing with adaptive window length to optimize the raw spectra and effectively remove noise.
• Dimensionality Reduction: Used Principal Component Analysis (PCA) to extract essential information from the full-spectrum data.
• Modeling: Constructed a Ridge Regression prediction model based on the processed data.

Results demonstrated that the combination of "AWL-SG smoothing + PCA dimensionality reduction + Ridge Regression" yielded the best prediction performance, achieving a coefficient of determination (R²) of 0.9146 on the test set, thereby validating the feasibility of this approach.

03 Clear Application Value: Supporting Agricultural Testing

This method provides scientifically robust firmness data to support apple harvesting, grading, storage, and transportation, contributing to reduced loss and enhanced quality control.

04 Conclusion

The study confirms that the Optosky ATP2000H spectrometer, as a reliable data acquisition tool, combined with efficient chemometric algorithms, can construct high-performance non-destructive detection models for apple firmness. This solution offers an effective technical reference for quality control in agricultural products.

Reference

1. Wenjing Gao, Xue Cheng, Xiaohan Liu, Yusheng Han, Zhenhui Ren, Apple firmness detection method based on hyperspectral technology, Food Control, Volume 166, 2024, 110690, ISSN 0956-7135, //doi.org/10.1016/j.foodcont.2024.110690.

Optosky ATP2000 Series Spectrometers

The ATP2000 series represents Optosky's high-performance spectrometers, featuring high sensitivity, broad spectral range, and excellent stability. They are widely used in plasma diagnostics, environmental monitoring, biomedical applications, and materials analysis. We are committed to providing professional spectral measurement tools and system solutions for scientific research and industrial users.