Class Embodiment Autoencoder (CEAE) for classifying the botanical origins of honey

Phillips, Tessa; Abdulla, Waleed H.

doi:10.1109/ivcnz48456.2019.8961004

Cited by 6 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A Hyperspectral imaging approach has been developed for detecting the botanical origins of honey [17,23,24,26,27,29]. In [27] the UMF grade of Manuka honey was predicted with 89% accuracy.…”

Section: Existing Quality Assurance Techniquesmentioning

confidence: 99%

“…In [27] the UMF grade of Manuka honey was predicted with 89% accuracy. The botanical origins of 21 different types of honey were predicted with 90% accuracy in [29]. These approaches used a hyperspectral imaging system to capture the data detailed in [21] and used a class embodiment autoencoder (CEAE) and support vector machines (SVM) for classification.…”

Section: Existing Quality Assurance Techniquesmentioning

confidence: 99%

“…Suitable benchmark classifiers, including the k-nearest neighbour (KNN) classifier and SVMs using a linear and radial basis function (RBF) kernel, have been considered. The feature reduction techniques used are: principal component analysis (PCA), selecting the k best features based on feature ranking, and autoencoder techniques [26,29].…”

Section: Feature Reduction and Classificationmentioning

confidence: 99%

“…Only the high variance features are kept in this approach [39]. PCA has been a high-performing benchmark method in our previous work on honey botanical origins classification [26,27,29].…”

Section: Principal Component Analysis (Pca)mentioning

confidence: 99%

“…There are existing measures to detect overall quality in premium honey types, such as New Zealand (NZ) Manuka honey [7]. Hyperspectral imaging has recently been developed to detect the botanical origins and quality of pure honey [23,26,27,29]. This approach has not yet been thoroughly investigated for detecting adulterated honey.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A new honey adulteration detection approach using hyperspectral imaging and machine learning

Phillips

Abdulla

2022

Eur Food Res Technol

Self Cite

View full text Add to dashboard Cite

This paper develops a new approach to fraud detection in honey. Specifically, we examine adulterating honey with sugar and use hyperspectral imaging and machine learning techniques to detect adulteration. The main contributions of this paper are introducing a new feature smoothing technique to conform to the classification model used to detect the adulterated samples and the perpetration of an adulterated honey data set using hyperspectral imaging, which has been made available online for the first time. Above $$95\%$$ 95 % accuracy was achieved for binary adulteration detection and multi-class classification between different adulterant concentrations. The system developed in this paper can be used to prevent honey fraud as a reliable, low cost, data-driven solution.

show abstract

Section: Existing Quality Assurance Techniquesmentioning

confidence: 99%

Section: Existing Quality Assurance Techniquesmentioning

confidence: 99%

Section: Feature Reduction and Classificationmentioning

confidence: 99%

“…Only the high variance features are kept in this approach [39]. PCA has been a high-performing benchmark method in our previous work on honey botanical origins classification [26,27,29].…”

Section: Principal Component Analysis (Pca)mentioning

confidence: 99%