Improved Deep CNN with Parameter Initialization for Data Analysis of Near-Infrared Spectroscopy Sensors

Near-infrared (NIR) spectral sensors can deliver the spectral response of light absorbed by materials. Data analysis technology based on NIR sensors has been a useful tool for quality identification. In this paper, an improved deep convolutional neural network (CNN) with batch normalization and MSRA (Microsoft Research Asia) initialization is proposed to discriminate the tobacco cultivation regions using data collected from NIR sensors. The network structure is created with six convolutional layers and three full connection layers, and the learning rate is controlled by exponential attenuation method. One-dimensional kernel is applied as the convolution kernel to extract features. Meanwhile, the methods of L2 regularization and dropout are used to avoid the overfitting problem, which improve the generalization ability of the network. Experimental results show that the proposed deep network structure can effectively extract the complex characteristics inside the spectrum, which proves that it has excellent recognition performance on tobacco cultivation region discrimination, and it also demonstrates that the deep CNN is more suitable for information mining and analysis of big data.

[1]  Yi Wang,et al.  [Tobacco quality analysis of producing areas of Yunnan tobacco using near-infrared (NIR) spectrum]. , 2013, Guang pu xue yu guang pu fen xi = Guang pu.

[2]  Antonis Nikitakis,et al.  Tensor-Based Classification Models for Hyperspectral Data Analysis , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[3]  Di Wang,et al.  Automatic Prediction of leave Chemical Compositions based on NIR spectroscopy with Machine Learning , 2019, Int. J. Robotics Autom..

[4]  Rashid Mehmood,et al.  Smarter Traffic Prediction Using Big Data, In-Memory Computing, Deep Learning and GPUs , 2019, Sensors.

[5]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[6]  Michael S. Lew,et al.  Deep learning for visual understanding: A review , 2016, Neurocomputing.

[7]  Yuhua Qin,et al.  NIR models for predicting total sugar in tobacco for samples with different physical states , 2016 .

[8]  Jannat Yasmin,et al.  Classification Method for Viability Screening of Naturally Aged Watermelon Seeds Using FT-NIR Spectroscopy , 2019, Sensors.

[9]  S. Ekren,et al.  The contents of some major and trace elements for quality groups of Aegean Region tobaccos , 2011 .

[10]  Pil Un Kim,et al.  Non-Destructive Classification of Diversely Stained Capsicum annuum Seed Specimens of Different Cultivars Using Near-Infrared Imaging Based Optical Intensity Detection , 2018, Sensors.

[11]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[12]  Lin Xie,et al.  Support Vector Machine Optimized by Genetic Algorithm for Data Analysis of Near-Infrared Spectroscopy Sensors , 2018, Italian National Conference on Sensors.

[13]  Nikolaos Doulamis,et al.  Deep supervised learning for hyperspectral data classification through convolutional neural networks , 2015, 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[14]  Nanning Zheng,et al.  Action Recognition by an Attention-Aware Temporal Weighted Convolutional Neural Network , 2018, Sensors.

[15]  Shaikh Anowarul Fattah,et al.  Biometric Authentication Using CNN Features of Dorsal Vein Pattern Extracted from NIR Image , 2018, TENCON 2018 - 2018 IEEE Region 10 Conference.

[16]  Tao Dong,et al.  Spectral Analysis and Sensitive Waveband Determination Based on Nitrogen Detection of Different Soil Types Using Near Infrared Sensors , 2018, Sensors.

[17]  Kang Ryoung Park,et al.  Deep Residual CNN-Based Ocular Recognition Based on Rough Pupil Detection in the Images by NIR Camera Sensor , 2019, Sensors.

[18]  Simon X. Yang,et al.  Intelligent Control of Bulk Tobacco Curing Schedule Using LS-SVM- and ANFIS-Based Multi-Sensor Data Fusion Approaches , 2019, Sensors.

[19]  Geoffrey E. Hinton,et al.  On rectified linear units for speech processing , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[20]  Tien Dat Nguyen,et al.  Spoof Detection for Finger-Vein Recognition System Using NIR Camera , 2017, Sensors.

[21]  Yuan-yuan Chen,et al.  Feature selection based convolutional neural network pruning and its application in calibration modeling for NIR spectroscopy , 2019, Chemometrics and Intelligent Laboratory Systems.

[22]  Yue Huang,et al.  Determination of 27 chemical constituents in Chinese southwest tobacco by FT-NIR spectroscopy , 2012 .

[23]  Elena Marchiori,et al.  Convolutional neural networks for vibrational spectroscopic data analysis. , 2017, Analytica chimica acta.

[24]  Q. Su,et al.  [Research on the nonlinear model of near infrared spectroscopy and the total sugar of tobacco samples]. , 2004, Guang pu xue yu guang pu fen xi = Guang pu.

[25]  S LewMichael,et al.  Deep learning for visual understanding , 2016 .

[26]  Amy Loutfi,et al.  A review of unsupervised feature learning and deep learning for time-series modeling , 2014, Pattern Recognit. Lett..

[27]  Ezzeddine Zagrouba,et al.  Convolutional neural network-based multimodal image fusion via similarity learning in the shearlet domain , 2018, Neural Computing and Applications.

[28]  Kang Ryoung Park,et al.  Multimodal Biometric Recognition Based on Convolutional Neural Network by the Fusion of Finger-Vein and Finger Shape Using Near-Infrared (NIR) Camera Sensor , 2018, Sensors.

[29]  Yizeng Liang,et al.  A modified random forest approach to improve multi-class classification performance of tobacco leaf grades coupled with NIR spectroscopy , 2016 .

[30]  Tom Fearn,et al.  Modern practical convolutional neural networks for multivariate regression: Applications to NIR calibration , 2018, Chemometrics and Intelligent Laboratory Systems.

[31]  Jens Meiler,et al.  Improving quantitative structure–activity relationship models using Artificial Neural Networks trained with dropout , 2016, Journal of Computer-Aided Molecular Design.

[32]  J. C. Dias,et al.  Fast inline tobacco classification by near-infrared hyperspectral imaging and support vector machine-discriminant analysis , 2019, Analytical Methods.

[33]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[34]  Kai Xie,et al.  Voiceprint Identification for Limited Dataset Using the Deep Migration Hybrid Model Based on Transfer Learning , 2018, Sensors.

[35]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[36]  Lei Chen,et al.  FPGA-Based Hybrid-Type Implementation of Quantized Neural Networks for Remote Sensing Applications , 2019, Sensors.

[37]  Nikolaos Doulamis,et al.  Deep Learning for Computer Vision: A Brief Review , 2018, Comput. Intell. Neurosci..

[38]  Tara N. Sainath,et al.  Improving deep neural networks for LVCSR using rectified linear units and dropout , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[39]  Kang Ryoung Park,et al.  Convolutional Neural Network-Based Classification of Driver’s Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors , 2018, Sensors.

[40]  Jin Wang,et al.  The Novel Sensor Network Structure for Classification Processing Based on the Machine Learning Method of the ACGAN , 2019, Sensors.