研究目的
To evaluate the use of NIR spectroscopy for non-destructive prediction of papaya maturity based on changes in water content, protein, and soluble solid content during harvest time.
研究成果
NIR spectroscopy combined with PLS models effectively predicts papaya maturity non-destructively, with high accuracy (R2 values of 0.85 for water content, 0.81 for protein, and 0.90 for soluble solid content). This technique can be adopted for determining optimal harvest time, improving fruit quality and storage management. Future studies should explore larger sample sizes and additional cultivars to enhance robustness.
研究不足
The study relies on a specific papaya cultivar ('IPB 1') and may not generalize to other varieties. The sample size (64 fruits) is relatively small, and the models are based on pre-defined harvest dates determined by operator experience, which could introduce subjectivity. The NIR spectroscopy method requires calibration and may be sensitive to environmental conditions and sample heterogeneity.
1：Experimental Design and Method Selection:
The study used NIR spectroscopy in reflection mode with a spectral range of 1000-2400 nm to acquire spectra from papaya samples. A partial least squares (PLS) algorithm was applied with pre-treatment processing (normalization, smoothing, and second derivative) to develop predictive models for water content, protein, and soluble solid content.
2：Sample Selection and Data Sources:
64 samples of papaya 'IPB 1' were harvested from a farmer's orchard at 14, 7, 4, and 0 days before the commercial harvest date. Quality parameters (water content, protein, soluble solid content) were measured destructively in triplicate.
3：List of Experimental Equipment and Materials:
NIRFlex N-500 spectrometer (Büchi Labortechnik AG), NIRWare
4：2 software, NIRCal 2 software, papaya fruits. Experimental Procedures and Operational Workflow:
Spectra were collected at room temperature (25°C) in triplicate per sample. Data were split into calibration (124 spectra) and validation (62 spectra) sets. Pre-treatment included normalization, smoothing, and second derivative. PLS models were built and validated using R2, SEC, and SEP metrics.
5：Data Analysis Methods:
Statistical analysis involved calculating coefficient of determination (R2), standard error of calibration (SEC), and standard error of prediction (SEP) to evaluate model accuracy and avoid over-fitting.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
