研究目的
To develop an optimal model to detect moisture content in peanut kernels based on hyperspectral imaging technique.
研究成果
The combination of chemometrics and hyperspectral imaging technology can achieve rapid and nondestructive detection of moisture content in peanut kernels. SPA–SVR was identified as the best model for this purpose.
研究不足
The study focused on four varieties of peanuts and used specific hyperspectral imaging equipment, which may limit the generalizability of the results to other varieties or equipment.
1：Experimental Design and Method Selection:
Hyperspectral imaging technology at 416–1000 nm was used to acquire hyperspectral images of peanut kernels. Three models (PLSR, PCR, and SVR) were established based on full wavelengths and effective wavelengths selected by SPA and RC.
2：Sample Selection and Data Sources:
Four varieties of peanuts were used as samples. Hyperspectral images were acquired using a 'push-broom' system.
3：List of Experimental Equipment and Materials:
High-performance electron multiplying CCD camera, line-scan imaging spectrograph, halogen light source equipment, one-axis electric linear mobile platform, and a Dell computer.
4：Experimental Procedures and Operational Workflow:
Peanut kernels were scanned to acquire hyperspectral images, which were then corrected and analyzed to extract spectral data.
5：Data Analysis Methods:
PLSR, PCR, and SVR models were used for data analysis, with performance measured by R2, RMSE, and RPD.
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
