Deep Spectral-spatial Features of Snapshot Hyperspectral Images for Red-meat Classification

We investigate the potential and accuracy of snapshot hyperspectral imaging for authentication and classification of red-meat species. Snapshot hyperspectral images are acquired of lamb, beef, and pork samples. We consider 13 muscles types of standard loin and leg chops. We propose a deep 3D convolution neural network (CNN) architecture for extracting and classifying spectral-spatial learned features of red-meat. We present a comparison with state-of-the-art models including partial least-square discriminant analysis and support vector machines. Our results show that the proposed 3D-CNN model outperforms the state-of-the-art models with 95.81% and 96.01% for overall accuracy and average F1score, respectively. Visualization results show that the proposed 3D-CNN model is able to convert snapshot hyperspectral image data into an intelligent representation with accurate separation between red-meat types. This study opens the door for more research towards real-time and completely portable hyperspectral imaging systems due to the ability of snapshot hyperspectral cameras to work at video rate.

[1]  Da-Wen Sun,et al.  Potential of hyperspectral imaging and pattern recognition for categorization and authentication of red meat , 2012 .

[2]  Xiuping Jia,et al.  Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural Networks , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[3]  Fan Zhang,et al.  Deep Convolutional Neural Networks for Hyperspectral Image Classification , 2015, J. Sensors.

[4]  Andy Lambrechts,et al.  A compact snapshot multispectral imager with a monolithically integrated per-pixel filter mosaic , 2014, Photonics West - Micro and Nano Fabricated Electromechanical and Optical Components.

[5]  Di Wu,et al.  Potential of time series-hyperspectral imaging (TS-HSI) for non-invasive determination of microbial spoilage of salmon flesh. , 2013, Talanta.

[6]  Di Wu,et al.  Application of visible and near infrared hyperspectral imaging for non-invasively measuring distribution of water-holding capacity in salmon flesh. , 2013, Talanta.

[7]  Gamal ElMasry,et al.  Application of NIR hyperspectral imaging for discrimination of lamb muscles , 2011 .

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

[9]  Baskar Ganapathysubramanian,et al.  Explaining hyperspectral imaging based plant disease identification: 3D CNN and saliency maps , 2018, ArXiv.

[10]  Guolan Lu,et al.  Medical hyperspectral imaging: a review , 2014, Journal of biomedical optics.

[11]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[12]  Dietrich Paulus,et al.  Hyperspectral Terrain Classification for Ground Vehicles , 2017, VISIGRAPP.

[13]  Di Wu,et al.  Study on infrared spectroscopy technique for fast measurement of protein content in milk powder based on LS-SVM , 2008 .

[14]  Pengcheng Nie,et al.  Determination of Calcium Content in Powdered Milk Using Near and Mid-Infrared Spectroscopy with Variable Selection and Chemometrics , 2012, Food and Bioprocess Technology.

[15]  H. Swatland.  Meat Cuts and Muscle Foods: An International Glossary , 2004 .

[16]  Dietrich Paulus,et al.  Robust Features for Snapshot Hyperspectral Terrain-Classification , 2017, CAIP.

[17]  Ying Li,et al.  Spectral-Spatial Classification of Hyperspectral Imagery with 3D Convolutional Neural Network , 2017, Remote. Sens..

[18]  Renfu Lu,et al.  Original paper: Hyperspectral waveband selection for internal defect detection of pickling cucumbers and whole pickles , 2010 .

[19]  Reinhard Klette,et al.  Concise Computer Vision , 2014, Undergraduate Topics in Computer Science.

[20]  Yong He,et al.  Theory and application of near infrared reflectance spectroscopy in determination of food quality , 2007 .

[21]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[22]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[23]  Reinhard Klette,et al.  Detection of Adulteration in Red Meat Species Using Hyperspectral Imaging , 2017, PSIVT.

[24]  Shigeki Nakauchi,et al.  Image analysis operations applied to hyperspectral images for non-invasive sensing of food quality – A comprehensive review , 2016 .

[25]  Reinhard Klette,et al.  Detection of Red-Meat Adulteration by Deep Spectral-Spatial Features in Hyperspectral Images , 2018, J. Imaging.

[26]  Panagiotis Tsakalides,et al.  Deep Convolutional Neural Networks for the Classification of Snapshot Mosaic Hyperspectral Imagery , 2017, Computational Imaging.

[27]  Raul Morais,et al.  Hyperspectral Imaging: A Review on UAV-Based Sensors, Data Processing and Applications for Agriculture and Forestry , 2017, Remote. Sens..

[28]  Junichi Kurihara,et al.  A novel approach for vegetation classification using UAV-based hyperspectral imaging , 2018, Comput. Electron. Agric..

[29]  Junichi Sugiyama,et al.  Foreign Substance Detection in Blueberry Fruits by Spectral Imaging , 2006 .

[30]  Reinhard Klette,et al.  Chemometrics and hyperspectral imaging applied to assessment of chemical, textural and structural characteristics of meat. , 2018, Meat science.