Session

Agriculture, Food Science and Technology

Description

The increasingly normative severity and market competitiveness have led the agriculture sector and the food industry to constantly look for logic improvements that can be applied in processes monitoring systems. In a context where fast, non-destructive and reliable techniques are required, image analysis-based methods have gained interest, thanks to their ability to spatially characterize heterogeneous samples. In such a scenario, HyperSpectral Imaging (HSI) is an emerging technique that provides not only spatial information of imaging systems, but even spectral information of spectroscopy. The utilization of the HSI approach opens new interesting scenario to quality control logics in agricultural and food processing/manufacturing sectors.

Three different case studies are presented in this paper. In particular, the utilization of an HSI system, working in SWIR range, was applied for: i) detecting contaminants in dried fruits to be packaged, ii) identifying olive fruits attacked by olive fruit flies and iii) recognizing flour type.

In particular, the proposed approach is based on the application of Partial Least Squares – Discriminant Analysis (PLS-DA) classification method to HyperSpectral images in Short Wave InfraRed (SWIR) range (1000-2500 nm). The proposed case studies demonstrate that this logic can be successfully utilized as a quality control system on agri-food products coming from different manufacturing stages, but it can even be seen as an analytical core for sorting engines.

Keywords:

hyperspectral imaging, agri-food products, quality control

Session Chair

Kastriot Pehlivani

Session Co-Chair

Violeta Lajqi Makolli

Proceedings Editor

Edmond Hajrizi

ISBN

978-9951-550-19-2

Location

Pristina, Kosovo

Start Date

26-10-2019 11:00 AM

End Date

26-10-2019 12:30 PM

DOI

10.33107/ubt-ic.2019.392

Share

COinS
 
Oct 26th, 11:00 AM Oct 26th, 12:30 PM

Hyperspectral imaging logics: efficient strategies for agri-food products quality control

Pristina, Kosovo

The increasingly normative severity and market competitiveness have led the agriculture sector and the food industry to constantly look for logic improvements that can be applied in processes monitoring systems. In a context where fast, non-destructive and reliable techniques are required, image analysis-based methods have gained interest, thanks to their ability to spatially characterize heterogeneous samples. In such a scenario, HyperSpectral Imaging (HSI) is an emerging technique that provides not only spatial information of imaging systems, but even spectral information of spectroscopy. The utilization of the HSI approach opens new interesting scenario to quality control logics in agricultural and food processing/manufacturing sectors.

Three different case studies are presented in this paper. In particular, the utilization of an HSI system, working in SWIR range, was applied for: i) detecting contaminants in dried fruits to be packaged, ii) identifying olive fruits attacked by olive fruit flies and iii) recognizing flour type.

In particular, the proposed approach is based on the application of Partial Least Squares – Discriminant Analysis (PLS-DA) classification method to HyperSpectral images in Short Wave InfraRed (SWIR) range (1000-2500 nm). The proposed case studies demonstrate that this logic can be successfully utilized as a quality control system on agri-food products coming from different manufacturing stages, but it can even be seen as an analytical core for sorting engines.