Potential of deep learning and snapshot hyperspectral imaging for classification of species in meat

Al-Sarayreh, Mahmoud; Reis, Marlon M.; Yan, Wei Qi; Klette, Reinhard

doi:10.1016/j.foodcont.2020.107332

Cited by 86 publications

(40 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although CNN models have been successfully implemented for remote sensing applications, they are not often applied to HSI data of food products. Earlier this year, Al-Sarayreh, et al [ 20 ] reported that 3D-CNN model approach applied to HSI data significantly enhanced the overall accuracy of red meat classification.…”

Section: Introductionmentioning

confidence: 99%

Deep Spectral-Spatial Features of Near Infrared Hyperspectral Images for Pixel-Wise Classification of Food Products

Zhu

Gowen

Feng³

et al. 2020

Sensors

View full text Add to dashboard Cite

Hyperspectral imaging (HSI) emerges as a non-destructive and rapid analytical tool for assessing food quality, safety, and authenticity. This work aims to investigate the potential of combining the spectral and spatial features of HSI data with the aid of deep learning approach for the pixel-wise classification of food products. We applied two strategies for extracting spatial-spectral features: (1) directly applying three-dimensional convolution neural network (3-D CNN) model; (2) first performing principal component analysis (PCA) and then developing 2-D CNN model from the first few PCs. These two methods were compared in terms of efficiency and accuracy, exemplified through two case studies, i.e., classification of four sweet products and differentiation between white stripe (“myocommata”) and red muscle (“myotome”) pixels on salmon fillets. Results showed that combining spectral-spatial features significantly enhanced the overall accuracy for sweet dataset, compared to partial least square discriminant analysis (PLSDA) and support vector machine (SVM). Results also demonstrated that spectral pre-processing techniques prior to CNN model development can enhance the classification performance. This work will open the door for more research in the area of practical applications in food industry.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Spectral-Spatial Features of Near Infrared Hyperspectral Images for Pixel-Wise Classification of Food Products

Zhu

Gowen

Feng³

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…In future, a more powerful backbone and training platform should be accommodated to design a better model [ 31 , 32 ]. This will assist us to design and compare those models more quickly and easily for fruits and food science [ 35 – 40 ]. In future, we will apply a newer model to achieve a better result for this fruit objection problem [ 33 , 34 , 41 ].…”

Section: Discussionmentioning

confidence: 99%

Fruit Detection from Digital Images Using CenterNet

Zhao

Yan

2021

Communications in Computer and Information Science

Self Cite

View full text Add to dashboard Cite

In this paper, CenterNet is chosen as the model to settle fruit detection problem from digital images. Three CenterNet models with various backbones were implemented, namely, ResNet-18, DLA-34, and Hourglass. A fruit dataset with four classes and 1,690 images was established for this research project. By comparing those models, followed the experimental results, the deep learning-based model with DLA-34 was selected as the final model to detect fruits from digital images, the performance is excellent. In this paper, the contribution is that we deploy a model based on CenterNet for visual object detection to resolve the problem of fruit detection. Meanwhile, there are 1,690 images distributed in four classes. Throughout evaluating the performance of the model, we eventually affirm the CenterNet based on DLA-34 to detect multiclass fruits from our images. The performance of this method is better than the existing ones in fruit detection.

show abstract

“…Application of NIR spectroscopy combined with PCA and PLS-DA for efficient discrimination and reliable detection of fraud in minced lamb and beef production was also reported [172]. Moreover, the use of NIR and deep 3D convolution neural network (3D-CNN) for red meat classification was demonstrated with a remarkable accuracy (>96.9%) [173].…”

Section: Meat and Porkmentioning

confidence: 97%

QCM Sensor Arrays, Electroanalytical Techniques and NIR Spectroscopy Coupled to Multivariate Analysis for Quality Assessment of Food Products, Raw Materials, Ingredients and Foodborne Pathogen Detection: Challenges and Breakthroughs

Bwambok

Siraj

Macchi

et al. 2020

Sensors

View full text Add to dashboard Cite

Quality checks, assessments, and the assurance of food products, raw materials, and food ingredients is critically important to ensure the safeguard of foods of high quality for safety and public health. Nevertheless, quality checks, assessments, and the assurance of food products along distribution and supply chains is impacted by various challenges. For instance, the development of portable, sensitive, low-cost, and robust instrumentation that is capable of real-time, accurate, and sensitive analysis, quality checks, assessments, and the assurance of food products in the field and/or in the production line in a food manufacturing industry is a major technological and analytical challenge. Other significant challenges include analytical method development, method validation strategies, and the non-availability of reference materials and/or standards for emerging food contaminants. The simplicity, portability, non-invasive, non-destructive properties, and low-cost of NIR spectrometers, make them appealing and desirable instruments of choice for rapid quality checks, assessments and assurances of food products, raw materials, and ingredients. This review article surveys literature and examines current challenges and breakthroughs in quality checks and the assessment of a variety of food products, raw materials, and ingredients. Specifically, recent technological innovations and notable advances in quartz crystal microbalances (QCM), electroanalytical techniques, and near infrared (NIR) spectroscopic instrument development in the quality assessment of selected food products, and the analysis of food raw materials and ingredients for foodborne pathogen detection between January 2019 and July 2020 are highlighted. In addition, chemometric approaches and multivariate analyses of spectral data for NIR instrumental calibration and sample analyses for quality assessments and assurances of selected food products and electrochemical methods for foodborne pathogen detection are discussed. Moreover, this review provides insight into the future trajectory of innovative technological developments in QCM, electroanalytical techniques, NIR spectroscopy, and multivariate analyses relating to general applications for the quality assessment of food products.

show abstract

Potential of deep learning and snapshot hyperspectral imaging for classification of species in meat

Cited by 86 publications

References 30 publications

Deep Spectral-Spatial Features of Near Infrared Hyperspectral Images for Pixel-Wise Classification of Food Products

Deep Spectral-Spatial Features of Near Infrared Hyperspectral Images for Pixel-Wise Classification of Food Products

Fruit Detection from Digital Images Using CenterNet

QCM Sensor Arrays, Electroanalytical Techniques and NIR Spectroscopy Coupled to Multivariate Analysis for Quality Assessment of Food Products, Raw Materials, Ingredients and Foodborne Pathogen Detection: Challenges and Breakthroughs

Contact Info

Product

Resources

About