A quality detection method of corn based on spectral technology and deep learning model.

Area of Science:

• Agricultural Science
• Spectroscopy
• Machine Learning

Background:

• Corn quality is crucial for global food security, breeding, and cultivation.
• Current corn quality detection methods are often cumbersome, slow, and inaccurate.

Purpose of the Study:

• To develop an efficient and accurate corn quality detection model.
• To integrate near-infrared (NIR) spectroscopy with deep learning for enhanced analysis.

Main Methods:

• Utilized near-infrared (NIR) spectroscopy combined with a deep learning convolutional neural network (LeNet-5).
• Applied wavelet transform (WT) and multivariate scattering correction (MSC) for spectral preprocessing.
• Employed the Competitive Adaptive Reweighted Sampling Algorithm (CARS) for feature selection.

Main Results:

• Achieved an average detection accuracy of 96.46% on the test set.
• Demonstrated superior performance over traditional machine learning models, with an average accuracy increase of 39.32%.

Conclusions:

• The proposed LeNet-5 deep learning model offers a highly accurate and efficient solution for corn quality detection.
• This approach overcomes limitations of traditional methods, paving the way for improved agricultural practices.