Moldy peanuts identification based on hyperspectral images and Point-centered convolutional neural network combined with embedded feature selection

Yuan, Deshuai; Jiang, Jinbao; Gong, Ziqiang; Nie, Cheng; Sun, Yi

doi:10.1016/j.compag.2022.106963

Cited by 22 publications

(8 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To verify the superiority of the 3D–2D hybrid CNN, the 43 feature bands (Y–Net(band)) extracted from the band selection mode of Y–Net were fed into CNN–ATT [ 25 ] and PCNN [ 24 ] for comparison with Y–Net(w) in this study.…”

Section: Resultsmentioning

confidence: 99%

“…Moreover, a large, trained network is reduced to a smaller one, enabling neural networks to be deployed in resource–constrained environments. In contrast to the two studies in [ 24 , 25 ], the authors used the network model for feature band selection and then retrained the model with the selected bands, and both the retrained models have reduced classification accuracy compared to the full–band models. So, it is good to train directly with the full band and then perform network pruning.…”

Section: Resultsmentioning

confidence: 99%

“…Therefore, some scholars have taken advantage of the CNN model to extract features by combining the selection of hyperspectral image feature bands with the CNN model. Yuan et al [ 24 ] integrated feature selection, extraction, and classification into an end–to–end trainable network. A point–centered convolutional neural network combined with an embedded feature selection network (PCNN–FS) is proposed to classify hyperspectral images of healthy and aflatoxin–infected moldy peanuts.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Y–Net: Identification of Typical Diseases of Corn Leaves Using a 3D–2D Hybrid CNN Model Combined with a Hyperspectral Image Band Selection Module

Jia

Shi

Luo

et al. 2023

Sensors

View full text Add to dashboard Cite

Corn diseases are one of the significant constraints to high–quality corn production, and accurate identification of corn diseases is of great importance for precise disease control. Corn anthracnose and brown spot are typical diseases of corn, and the early symptoms of the two diseases are similar, which can be easily misidentified by the naked eye. In this paper, to address the above problems, a three–dimensional–two–dimensional (3D–2D) hybrid convolutional neural network (CNN) model combining a band selection module is proposed based on hyperspectral image data, which combines band selection, attention mechanism, spatial–spectral feature extraction, and classification into a unified optimization process. The model first inputs hyperspectral images to both the band selection module and the attention mechanism module and then sums the outputs of the two modules as inputs to a 3D–2D hybrid CNN, resulting in a Y–shaped architecture named Y–Net. The results show that the spectral bands selected by the band selection module of Y–Net achieve more reliable classification performance than traditional feature selection methods. Y–Net obtained the best classification accuracy compared to support vector machines, one–dimensional (1D) CNNs, and two–dimensional (2D) CNNs. After the network pruned the trained Y–Net, the model size was reduced to one–third of the original size, and the accuracy rate reached 98.34%. The study results can provide new ideas and references for disease identification of corn and other crops.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Y–Net: Identification of Typical Diseases of Corn Leaves Using a 3D–2D Hybrid CNN Model Combined with a Hyperspectral Image Band Selection Module

Jia

Shi

Luo

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…CNN is a deep learning method with structural diversity and nonlinear transformation that has achieved significant achievements in many fields such as image processing, speech recognition, and text data [20,24]. In recent years, CNN has been intensively researched and explored in spectral analysis and extended to applications for one-dimensional (1D) data (such as pixel-level spectra) and three-dimensional (3D) data [43].…”

Section: Convolutional Neural Network Architecture For Feature Select...mentioning

confidence: 99%

“…With the rapid development of deep learning techniques, convolutional neural network (CNN)based methods have been successfully applied to hyperspectral band selection [19]. Yuan proposed a point-centered convolutional neural network incorporating embedded feature selection for feature selection, extraction, and classification, and the accuracy of the selected five critical bands reached 97.98% for non-destructive and rapid identification of moldy peanuts [20]. Sharma proposed a CNN named DeepFeature applied to non-image data for feature selection with 98% classification accuracy on an independent test set, which can provide a powerful method for identifying biologically relevant gene sets [21].…”

Section: Introductionmentioning

confidence: 99%

Identification of Defective Maize Seeds Using Hyperspectral Imaging Combined with Deep Learning

Sun

et al. 2022

Foods

View full text Add to dashboard Cite

Seed quality affects crop yield and the quality of agricultural products, and traditional identification methods are time-consuming, complex, and irreversibly destructive. This study aims to establish a fast, non-destructive, and effective approach for defect detection in maize seeds based on hyperspectral imaging (HSI) technology combined with deep learning. Raw spectra collected from maize seeds (200 each healthy and worm-eaten) were pre-processed using detrending (DE) and multiple scattering correction (MSC) to highlight the spectral differences between samples. A convolutional neural network architecture (CNN-FES) based on a feature selection mechanism was proposed according to the importance of wavelength in the target classification task. The results show that the subset of 24 feature wavelengths selected by the proposed CNN-FES can capture important feature information in the spectral data more effectively than the conventional successive projections algorithm (SPA) and competitive adaptive reweighted sampling (CARS) algorithms. In addition, a convolutional neural network architecture (CNN-ATM) based on an attentional classification mechanism was designed for one-dimensional spectral data classification and compared with three commonly used machine learning methods, linear discriminant analysis (LDA), random forest (RF), and support vector machine (SVM). The results show that the classification performance of the designed CNN-ATM on the full wavelength does not differ much from the above three methods, and the classification accuracy is above 90% on both the training and test sets. Meanwhile, the accuracy, sensitivity, and specificity of CNN-ATM based on feature wavelength modeling can reach up to 97.50%, 98.28%, and 96.77% at the highest, respectively. The study shows that hyperspectral imaging-based defect detection of maize seed is feasible and effective, and the proposed method has great potential for the processing and analysis of complex hyperspectral data.

show abstract

Embedded Temporal Feature Selection for Time Series Forecasting Using Deep Learning

Jiménez-Navarro,

Martínez-Ballesteros,

Martínez-Álvarez

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Moldy peanuts identification based on hyperspectral images and Point-centered convolutional neural network combined with embedded feature selection

Cited by 22 publications

References 42 publications

Y–Net: Identification of Typical Diseases of Corn Leaves Using a 3D–2D Hybrid CNN Model Combined with a Hyperspectral Image Band Selection Module

Y–Net: Identification of Typical Diseases of Corn Leaves Using a 3D–2D Hybrid CNN Model Combined with a Hyperspectral Image Band Selection Module

Identification of Defective Maize Seeds Using Hyperspectral Imaging Combined with Deep Learning

Embedded Temporal Feature Selection for Time Series Forecasting Using Deep Learning

Contact Info

Product

Resources

About