Abstract

A global predictive model was developed for protein, moisture, and grain type, using near infrared (NIR) spectra. The model is a deep convolutional neural network, trained on NIR spectral data captured from wheat, barley, field pea, and lentil whole grains. The deep learning model performs multi-task learning to simultaneously predict grain protein, moisture, and type, with a significant reduction in prediction errors compared to linear approaches (e.g., partial least squares regression). Moreover, it is shown that the convolutional network architecture learns much more efficiently than simple feedforward neural network architectures of the same size. Thus, in addition to improved accuracy, the presented deep network is very efficient to implement, both in terms of model development time, and the required computational resources.

© 2020 The Author(s)

Single-model multi-tasks deep learning network for recognition and quantitation of surface-enhanced Raman spectroscopy

Luyuan Xie, Yujian Shen, Meng Zhang, Yan Zhong, Yi Lu, Longkun Yang, and Zhipeng Li
Opt. Express 30(23) 41580-41589 (2022)

End-to-end algorithm for the automatic detection of the neural canal opening in OCT images based on a multi-task deep learning model

Chieh-En Lee, Jia-Ling Tu, Pei-Chia Tsai, Yu-Chieh Ko, Shih-Jen Chen, Ying-Shan Chen, Chu-Ming Cheng, and Chung-Hao Tien
Opt. Continuum 2(9) 2055-2068 (2023)

CGAN-rIRN: a data-augmented deep learning approach to accurate classification of mental tasks for a fNIRS-based brain-computer interface

Yao Zhang, Dongyuan Liu, Tieni Li, Pengrui Zhang, Zhiyong Li, and Feng Gao
Biomed. Opt. Express 14(6) 2934-2954 (2023)

Intelligent optical performance monitor using multi-task learning based artificial neural network

Zhiquan Wan, Zhenming Yu, Liang Shu, Yilun Zhao, Haojie Zhang, and Kun Xu
Opt. Express 27(8) 11281-11291 (2019)

Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces

Tevfik Bulent Kanmaz, Efe Ozturk, Hilmi Volkan Demir, and Cigdem Gunduz-Demir
Optica 10(10) 1373-1382 (2023)