English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON NEAR INFRARED SPECTROSCOPY FOZ DO IGUASSU BRAZIL 18-23 OCTOBER 2015 IntroductIon Hyperspectral imaging (HSI) is a non-destructive technique to explore surfaces in more detail than single point spectroscopy. Each pixel of a hyperspectral image, usually highly correlated to its neighbours, contains the spectrum of that specific position. Thus, the hyperspectral image contains both spatial and spectral information of a sample. The data are organized in three-dimensional blocks, called hypercubes. Many devices for acquiring hyperspectral images have been manufactured and there is an increasing interest for improving the data analysis techniques applied to such complex datasets1. Despite being one of the main advantages of hyperspectral systems, the large datasets routinely encountered in HSI, can complicate the extraction of useful information since much of the information obtained is redundant2. Multivariate data analysis, or Chemometrics, is a suitable approach to reduce the dimensionality of the data while retaining the most useful spectral information. A useful methodology that can reduce the data dimension is the variable selection. In variable selection wavelengths that would be most influential on product quality evaluation are selected, removing the wavelengths that have no or low discrimination power. In this way, the data dimension is reduced while preserving the most useful information3.The variables selected depend on the behaviour of spectral responses under modification of the samples and on differences among them. Hyperspectral imaging (HSI) is a powerful analytical tool but the large amount of data typically generated during HSI acquisition can limit subsequent industrial applications. In order to extract the relevant information, it is possible to adapt the classical chemometric techniques for the identification of key wavelengths for a given quality control issue. This would allow the development of multispectral devices, intended for specific product control. Acerola is a typical Brazilian fruit that at the initial stages of ripening is green, changing to yellow-reddish colour and finally to purple when it is completely ripened. Due to its high moisture content, a rapid deterioration is commonly observed on ripening. This work is focused on the changes in post-harvest quality of Acerola fruits, with the aim of developing an easy way to visualize the global fruit status during ripening. An investigation of 20 samples for five consecutive days was carried out to highlight the most important wavelengths that characterized the maturity/senescence process of the Acerola fruit, investigating the evolution of acerolas mean spectra during time. The three wavelengths selected were 1883 nm, 1407 nm and 1136 nm, related to absorption and overtones of OH and CH bonds. Grey scale images at these selected wavelengths were combined to provide a false RGB image that allowed evaluation of the ripening process in a rapid and non-destructive manner. These results will facilitate the development of a low cost multispectral imaging system characterized by a simple image based output that could improve quality monitoring of acerola.

Selection of NIR wavelengths from hyperspectral imaging data for the quality evaluation of Acerola fruit