On-line adaptive Bayesian classification for in-line particle image monitoring in polymer film manufacturing

Torabi, Keivan; Sayad, Saed; Balke, S. T.

doi:10.1016/j.compchemeng.2005.06.008

Cited by 9 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Trials using images from in-line monitoring of polyethylene flowing in an extruder provided particle count results equivalent to those of a human observer and particle size accuracies limited mostly by the resolution of the camera system. Very recently the MaxMin thresholding method has been successfully used to provide data for in-line classification of images into those without particles and those with particles (Torabi et al, 2005). In that work, the method was applied to thousands of images of variable quality from in-line image monitoring.…”

Section: Discussionmentioning

confidence: 99%

Adaptive image thresholding for real‐time particle monitoring

Torabi

Sayad

Balke

2006

Int J Imaging Syst Tech

Self Cite

View full text Add to dashboard Cite

Image thresholding is critical to computer vision systems designed to detect very small numbers of contaminant particles from analysis of images acquired by in-line process monitoring. The objective of this work was to obtain a thresholding method that would permit in-line, \real-time," determination of both the number of particles in an image and their size. An additional requirement was that it automatically adapt to inevitable variations in the image quality. A new global image thresholding method, the MaxMin method (\MaxMin"), was developed. MaxMin notes the size of the smallest detected particle in an image as threshold value is progressively changed from black to white. The selected threshold value is the one providing the largest size. MaxMin was tested on thousands of images, and it was shown to readily adapt to images of different background noise levels and provided particle counts as accurate as those of a human observer in less than three seconds per image. Error in particle size measurement was a function of the particle size and the image resolution. It was about 3% for 50 m particles, using a CCD camera with 2Â lens, calibrated for each pixel to represent *5 m 2 . The error was significantly higher for smaller particles, when the same system resolution was used.

show abstract

Section: Discussionmentioning

confidence: 99%

Adaptive image thresholding for real‐time particle monitoring

Torabi

Sayad

Balke

2006

Int J Imaging Syst Tech

Self Cite

View full text Add to dashboard Cite

show abstract

“…The adaptive machine learning is not new and it also includes AutoML concept. The concept of adaptive machine learning can be dated back to 1990s (Blum, ; Littlestone & Warmuth, ), as stated in the paper (Torabi, Sayad, & Balke, ). Today, it can be observed that the big data characteristics and the current widespread interdisciplinary applications have enforced new constraints and requirements that triggered the exploration of novel approaches for adaptive machine learning.…”

Section: Smart Machine Learningmentioning

confidence: 99%

Big data analytics: Machine learning and Bayesian learning perspectives—What is done? What is not?

Suthaharan

2018

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Big data analytics provides an interdisciplinary framework that is essential to support the current trend for solving real‐world problems collaboratively. The progression of big data analytics framework must be clearly understood so that novel approaches can be developed to advance this state‐of‐the‐art discipline. An ignorance of observing the progression of this fast‐growing discipline may lead to duplications in research and waste of efforts. Its main companion field, machine learning, helps solve many big data analytics problems; therefore, it is also important to understand the progression of machine learning in the big data analytics framework. One of the current research efforts in big data analytics is the integration of deep learning and Bayesian optimization, which can help the automatic initialization and optimization of hyperparameters of deep learning and enhance the implementation of iterative algorithms in software. The hyperparameters include the weights used in deep learning, and the number of clusters in Bayesian mixture models that characterize data heterogeneity. The big data analytics research also requires computer systems and software that are capable of storing, retrieving, processing, and analyzing big data that are generally large, complex, heterogeneous, unstructured, unpredictable, and exposed to scalability problems. Therefore, it is appropriate to introduce a new research topic—transformative knowledge discovery—that provides a research ground to study and develop smart machine learning models and algorithms that are automatic, adaptive, and cognitive to address big data analytics problems and challenges. The new research domain will also create research opportunities to work on this interdisciplinary research space and develop solutions to support research in other disciplines that may not have expertise in the research area of big data analytics. For example, the research, such as detection and characterization of retinal diseases in medical sciences and the classification of highly interacting species in environmental sciences can benefit from the knowledge and expertise in big data analytics. This article is categorized under: Application Areas > Science and Technology Technologies > Machine Learning Fundamental Concepts of Data and Knowledge > Big Data Mining

show abstract