Raman spectrum classification based on transfer learning by a convolutional neural network: Application to pesticide detection

Hu, Jiaqi; Zou, Yanqiu; Sun, Biao; Yu, Xin‐Yao; Shang, Ziyang; Huang, Jie; Jin, Shangzhong; Liang, Pei

doi:10.1016/j.saa.2021.120366

Cited by 32 publications

(16 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, researchers have proposed using deep learning for target substance detection and identification in Raman spectra. 14,15 However, deep-learning models are data hungry and heavily rely on a substantial amount of labeled training data for accurate predictions. Acquiring a significant amount of actual Raman spectral data to adequately train the deep-learning model poses a challenge.…”

Section: ■ Introductionmentioning

confidence: 99%

Library-Based Raman Spectral Identification Using Multi-Input Hybrid ResNet

Chen,

Baek

2023

ACS Omega

View full text Add to dashboard Cite

Raman spectroscopy is widely used for its exceptional identification capabilities in various fields. Traditional methods for target identification using Raman spectroscopy rely on signal correlation with moving windows, requiring data preprocessing that can significantly impact identification performance. In recent years, deep-learning approaches have been proposed to leverage data augmentation techniques, such as baseline and additive noise addition, in order to overcome data scarcity. However, these deep-learning methods are limited to the spectra encountered during training and struggle to handle unseen spectra. To address these limitations, we propose a multi-input hybrid deep-learning model trained with simulated spectral data. By employing simulated spectra, our method tackles the challenges of data scarcity and the handling of unseen spectra encountered in traditional and deep-learning methods. Experimental results demonstrate that our proposed method achieves outstanding identification performance and effectively handles spectra obtained from different Raman spectroscopy systems.

show abstract

Section: ■ Introductionmentioning

confidence: 99%

Library-Based Raman Spectral Identification Using Multi-Input Hybrid ResNet

Chen,

Baek

2023

ACS Omega

View full text Add to dashboard Cite

show abstract

“…Furthermore, in real-life applications, obtaining and understanding Raman signals are challenging. For this purpose, supplementing Raman analysis with machine learning (ML) algorithms can automate detection and facilitate discrimination of pesticides. , …”

Section: Introductionmentioning

confidence: 99%

“…For this purpose, supplementing Raman analysis with machine learning (ML) algorithms can automate detection and facilitate discrimination of pesticides. 44,45 Here, we report an ML-assisted pesticide detection approach on an eco-friendly and low-cost SERS platform. To the best of our knowledge, this is the first study to evaluate the detection of water pollution agents by developing pollutant-free (including solvent), eco-friendly SERS substrates.…”

Section: ■ Introductionmentioning

confidence: 99%

Machine Learning-Assisted Pesticide Detection on a Flexible Surface-Enhanced Raman Scattering Substrate Prepared by Silver Nanoparticles

Şahin

Çelik

Çamdal

et al. 2022

ACS Appl. Nano Mater.

View full text Add to dashboard Cite

Access to clean water is a pressing challenge affecting millions of lives and the aquatic body of the Earth. Sensitive detection of pollutants such as pesticides is particularly important to address this challenge. This study reports eco-friendly preparation of the surface-enhanced Raman scattering (SERS) substrate for machine learning-assisted detection of pesticides in water. The proposed SERS platform was prepared on a copy paper by reducing a silver salt using the extract of a natural plant, Cedrus libani. The fabricated SERS platform was characterized in detail using scanning electron microscopy, energy-dispersive X-ray spectroscopy, X-ray diffraction, and X-ray photoelectron spectroscopy. The high-density formation of silver nanoparticles with an average diameter of 41 nm on the surface of the paper enabled detection of analytes with a nanomolar level sensitivity. This SERS capability was used to collect Raman signals of four different pesticides in water: myclobutanil, phosmet, thiram, and abamectin. Raman spectra of the pesticides are highly complex, challenging accurate determination of the pesticide type. To overcome this challenge and distinguish pesticides, machine learning (ML) approach was used. The ML-mediated detection of harmful pesticides on a versatile, green, and inexpensive SERS platform appears to be promising for environmental applications.

show abstract

“…On the other hand, transfer learning, which applied learned knowledge in previous tasks to new tasks, provides an easy way for the modeling problem. By transferring the weights of the pre-trained model to a new task, minor modifications are required to build the new model. , Thus, data requirements can be significantly reduced . Classification using transfer learning were reported, demonstrating that gasoline grade can be determined by fine-tuning the network, and the last three layers of the GoogleLeNet model are replaced with new layers, in which GoogleLeNet is pre-trained through classifying images in 1000 categories …”

Section: Introductionmentioning

confidence: 99%

Spectral Encoder to Extract the Features of Near-Infrared Spectra for Multivariate Calibration

Duan

Liu

Cai

et al. 2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

An autoencoder architecture was adopted for near-infrared (NIR) spectral analysis by extracting the common features in the spectra. Three autoencoder-based networks with different purposes were constructed. First, a spectral encoder was established by training the network with a set of spectra as the input. The features of the spectra can be encoded by the nodes in the bottleneck layer, which in turn can be used to build a sparse and robust model. Second, taking the spectra of one instrument as the input and that of another instrument as the reference output, the common features in both spectra can be obtained in the bottleneck layer. Therefore, in the prediction step, the spectral features of the second can be predicted by taking the reverse of the decoder as the encoder. Furthermore, transfer learning was used to build the model for the spectra of more instruments by fine-tuning the trained network. NIR datasets of plant, wheat, and pharmaceutical tablets measured on multiple instruments were used to test the method. The multi-linear regression (MLR) model with the encoded features was found to have a similar or slightly better performance in prediction compared with the partial least-squares (PLS) model.

show abstract

Raman spectrum classification based on transfer learning by a convolutional neural network: Application to pesticide detection

Cited by 32 publications

References 34 publications

Library-Based Raman Spectral Identification Using Multi-Input Hybrid ResNet

Library-Based Raman Spectral Identification Using Multi-Input Hybrid ResNet

Machine Learning-Assisted Pesticide Detection on a Flexible Surface-Enhanced Raman Scattering Substrate Prepared by Silver Nanoparticles

Spectral Encoder to Extract the Features of Near-Infrared Spectra for Multivariate Calibration

Contact Info

Product

Resources

About