Methodological approach for the automatic discrimination of pictorial materials using fused hyperspectral imaging data from the visible to mid-infrared range coupled with machine learning methods

Capobianco, G.; Pronti, Lucilla; Gorga, E.; Romani, M.; Cestelli-Guidi, M.; Serranti, Silvia; Bonifazi, G.

doi:10.1016/j.saa.2023.123412

Cited by 7 publications

(3 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different pre-processing strategies were tested following the combinations widely adopted in literature [28] . More in detail, spectral differences among the different samples have been emphasized and baseline background was removed, applying a Savitzky-Golay (SG) polynomial derivative filter.…”

Section: Acquired Hyperspectral Images Processingmentioning

confidence: 99%

“…More in detail, spectral differences among the different samples have been emphasized and baseline background was removed, applying a Savitzky-Golay (SG) polynomial derivative filter. Finally, before PCA all data was centered using mean center [28] .…”

Section: Acquired Hyperspectral Images Processingmentioning

confidence: 99%

“…In this scenario, HSI approach combined with chemometric strategies enables the evaluation of chemical attributes of the sample exposed surface, thanks to the already mentioned 3D nature (i.e., spectral and spatial) of hyperspectral data [17] . For these reasons, HSI applications have rapidly grown in different sectors, such as agri-food [18] , waste recycling [19][20] hazardous waste detection [21][22] , environmental monitoring [23][24] , medical [25][26] and cultural heritage [27][28] . The present work aims to develop an approach based on HSI in the SWIR range to enable winegrowers and technical advisers to rapidly determine the macro-and micronutrients in vine leaves in order to evaluate the physiology condition.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Monitoring of macro- and micronutrients in vine leaves by short-wave infrared hyperspectral imaging

Bonifazi,

Capobianco,

Serranti

et al. 2024

Photonic Technologies in Plant and Agricultural Science

Self Cite

View full text Add to dashboard Cite

Wine is a widely diffused beverage in the world and its production and quality are greatly influenced by the health conditions of the vine plants. The aim of the study was to investigate the possibility to monitor by hyperspectral imaging (HSI) the macronutrients (i.e., Ca, K, etc.) and micronutrients (i.e., Mn, Cu, Zn, etc.) variation in leaves sampled from different areas of a vineyard. The proposed approach is based on the acquisition by HSI in the short-wave infrared range (SWIR: 1000-2500 nm), of dried and milled vine leaves, followed by the implementation of a classification model based on Partial Least Square (PLS). Micro-X-ray fluorescence (micro-XRF) analyses were carried out on the same samples to correlate the SWIR spectral signatures with the detected chemical elements. Furthermore, HSI-based prediction maps, representative of the chemical elements distribution in samples were obtained. The achieved results are very promising, especially with reference to the possibility to adopt a fast strategy to monitor the macro-and micronutrients variation in the leaves directly in field, allowing to treat in real-time any nutritional deficiencies.

show abstract

Section: Acquired Hyperspectral Images Processingmentioning

confidence: 99%

Section: Acquired Hyperspectral Images Processingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Monitoring of macro- and micronutrients in vine leaves by short-wave infrared hyperspectral imaging

Bonifazi,

Capobianco,

Serranti

et al. 2024

Photonic Technologies in Plant and Agricultural Science

Self Cite

View full text Add to dashboard Cite

show abstract

Unveiling Thymoma Typing Through Hyperspectral Imaging and Deep Learning

Lv,

Liang,

Tian

et al. 2024

Journal of Biophotonics

View full text Add to dashboard Cite

Thymoma, a rare tumor from thymic epithelial cells, presents diagnostic challenges because of the subjective nature of traditional methods, leading to high false‐negative rates and long diagnosis times. This study introduces a thymoma classification technique that integrates hyperspectral imaging with deep learning. We initially capture pathological slice images of thymoma using a hyperspectral camera and delineate regions of interest to extract spectral data. This data undergoes reflectance calibration and noise reduction. Subsequently, we transform the spectral data into two‐dimensional images via the Gramian Angular Field (GAF) method. A variant residual network is then utilized to extract features and classify these images. Our results demonstrate that this model significantly enhances classification accuracy and efficiency, achieving an average accuracy of 95%. The method proves highly effective in automated thymoma diagnosis, optimizing data utilization, and feature representation learning.

show abstract