Unsupervised Clustering of Hyperspectral Paper Data Using t-SNE

Devassy, Binu Melit; George, Sony; Nussbaum, Peter

doi:10.3390/jimaging6050029

Cited by 48 publications

(43 citation statements)

References 33 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Through its process, t-SNE can detect clusters in data very well (Linderman & Steinerberger, 2019). Therefore, the results of t-SNE are suitable for identifying best clusters using traditional clustering techniques (Dhalmahapatra et al, 2019;Melit Devassy et al, 2020). This relationship was also observed in our data set.…”

Section: Appendixsupporting

confidence: 70%

The Structure of Social Networks and Its Link to Higher Education Students' Socio-Emotional Loneliness During COVID-19

Hopp¹,

Händel²,

Bedenlier³

et al. 2021

Preprint

View full text Add to dashboard Cite

Lonely students typically underperform academically. According to several studies, the COVID-19 pandemic is an important risk factor for increases in loneliness, as the contact restrictions and the switch to mainly online classes potentially burden the students. The previously familiar academic environment (campus) as well as the exchange with peers and lecturers on site were no longer made available. In our study, we examine factors that could potentially counteract the development of higher education student loneliness during the COVID-19 pandemic from a social network perspective. During the semester, N = 283 students from across all institutional faculties of a German comprehensive university took part in an online survey. We surveyed their social and emotional experiences of loneliness, their self-reported digital skills, and their current egocentric networks. We distinguished between close online contacts (i.e., mainly online exchanges) and close offline contacts (i.e., mainly in situ exchanges). In addition, we derived the interconnectedness (i.e., the densities of the egocentric networks) and diversity (operationalized with the entropy) of students’ contacts. The results of correlation analyses and hierarchical linear regressions indicate that strong digital skills are related to both a higher number of online contacts and to lower social and emotional experiences of loneliness. Regardless of whether offline or online, the number of reported contacts is indicative of a lower experience of social loneliness. A well-connected network related to lower experiences of social but not emotional loneliness. Finally, findings suggest that homogenous networks tend to be related with lower experiences of both social and emotional loneliness. Overall, our study indicates that barriers to online communication might be mitigating factors to consider when assessing the impact of the COVID-19 pandemic on student loneliness.

show abstract

Section: Appendixsupporting

confidence: 70%

The Structure of Social Networks and Its Link to Higher Education Students' Socio-Emotional Loneliness During COVID-19

Hopp¹,

Händel²,

Bedenlier³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The samples used in this study were already presented in our previous work, 39 which contains 40 commonly used paper types collected and arranged as a checkerboard pattern (10 rows Â 4 columns). Each sample has a square shape with 4 cm long sides, as shown in Figure 1.…”

Section: Samples and Hyperspectral Image Acquisitionmentioning

confidence: 99%

Classification of forensic hyperspectral paper data using hybrid spectral similarity algorithms

Deepthi

Devassy

George

et al. 2022

Journal of Chemometrics

Self Cite

View full text Add to dashboard Cite

Document forgeries that involve modification of the materials used, such as ink and paper, provide evidence of any malpractices being performed. Forensic specialists use different techniques to identify and classify these samples; however, the most preferred method is to use nondestructive techniques to avoid any potential damage to the original specimen under investigation. Hyperspectral imaging has already been explored in several application domains and used as a powerful method in forensic investigations to extract information about the materials under examination. To precisely classify the material information and utilize the hyperspectral imaging technique's potential, we probed the potential of some hybrid spectral similarity measures to classify different commonly used paper samples. A comparison of these methods is quantitatively presented in this article.

show abstract

“…HSI has received wide acceptance in food analysis as it offers the possibility to simultaneously record both spectral and spatial information. As shown in Figure 1, HSI data can be viewed as a data cube 17 that consists of different band images as layers and is often referred to as an HSI data cube. It has spatial information along the X-and Y-axes and spectral information along the Z-axis.…”

Section: Introductionmentioning

confidence: 99%

Estimation of strawberry firmness using hyperspectral imaging: a comparison of regression models

Devassy

George

2021

J. Spectral Imaging

Self Cite

View full text Add to dashboard Cite

Firmness is one of the most important quality measures of strawberries, and is related to other aspects of the fruit, such as flavour, ripeness and internal characteristics. The most popular method for measuring firmness is puncturing with a penetrometer, which is destructive and time-consuming. In the present study, we make an attempt to predict the firmness of strawberries in a fast, non-destructive and non-contact way using hyperspectral imaging (HSI) and data analysis with various regression techniques. The primary goal of this research is to investigate and compare the firmness prediction capability of seven prominent regression techniques. We have performed HSI data acquisition of 150 strawberries and optimised seven regression models using the spectral information to predict strawberry firmness. These models are linear, ridge, lasso, k-neighbours, random forest, support vector and partial least square regression. The res ults show that HSI data with regression models has the potential to predict firmness in a rapid, non-destructive manner. Out of these seven regression models, the k-neighbours regression model outperformed all other methods with a standard error of prediction of 0.14, which is better than that of the state-of-the-art results.

show abstract

Unsupervised Clustering of Hyperspectral Paper Data Using t-SNE

Cited by 48 publications

References 33 publications

The Structure of Social Networks and Its Link to Higher Education Students' Socio-Emotional Loneliness During COVID-19

The Structure of Social Networks and Its Link to Higher Education Students' Socio-Emotional Loneliness During COVID-19

Classification of forensic hyperspectral paper data using hybrid spectral similarity algorithms

Estimation of strawberry firmness using hyperspectral imaging: a comparison of regression models

Contact Info

Product

Resources

About