Automatic counting method for complex overlapping erythrocytes based on seed prediction in microscopic imaging

Wei, Xudong; Cao, Yiping

doi:10.1142/s1793545816500164

Cited by 6 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Python3 libraries such as NumPy, SciPy, scikit-learn, Keras, pandas, and Matplotlib are utilized to perform the categorization through ML models. Scikit-learn appears to be the most user-friendly and reliable machine learning library [49,50]. e foundations of this package are NumPy, SciPy, and Matplotlib.…”

Section: Methodsmentioning

confidence: 99%

Wheat Seed Classification: Utilizing Ensemble Machine Learning Approach

Khatri¹,

Agrawal²,

Chatterjee³

2022

Scientific Programming

View full text Add to dashboard Cite

Recognizing and authenticating wheat varieties is critical for quality evaluation in the grain supply chain, particularly for methods for seed inspection. Recognition and verification of grains are carried out manually through direct visual examination. Automatic categorization techniques based on machine learning and computer vision offered fast and high-throughput solutions. Even yet, categorization remains a complicated process at the varietal level. The paper utilized machine learning approaches for classifying wheat seeds. The seed classification is performed based on 7 physical features: area of wheat, perimeter of wheat, compactness, length of the kernel, width of the kernel, asymmetry coefficient, and kernel groove length. The dataset is collected from the UCI library and has 210 occurrences of wheat kernels. The dataset contains kernels from three wheat varieties Kama, Rosa, and Canadian, with 70 components chosen at random for the experiment. In the first phase, K-nearest neighbor, classification and regression tree, and Gaussian Naïve Bayes algorithms are implemented for classification. The results of these algorithms are compared with the ensemble approach of machine learning. The results reveal that accuracies calculated for KNN, decision, and Naïve Bayes classifiers are 92%, 94%, and 92%, respectively. The highest accuracy of 95% is achieved through the ensemble classifier in which decision is made based on hard voting.

show abstract

Section: Methodsmentioning

confidence: 99%

Wheat Seed Classification: Utilizing Ensemble Machine Learning Approach

Khatri¹,

Agrawal²,

Chatterjee³

2022

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…The procedure (Fig. 2) is inspired by the one described by Wei and Cao (2016), consisting in smoothing the image using a mean filter, then the application of a threshold separating the pores and the matrix, and finally the watershed treatment available in ImageJ in order to get the surface area and the coordinates of the centroid of each pore. The pores can be assimilated to perfect circles the diameters of which are calculated from Eq.…”

Section: Image Processingmentioning

confidence: 99%

Mechanical properties of a nanoporous membrane used in implantable medical devices. Correlation between experimental characterization and 2D numerical simulation

Cristofari

Piotrowski

Pesci

2017

Journal of the Mechanical Behavior of Biomedical Materials

View full text Add to dashboard Cite

is an open access repository that collects the work of Arts et Métiers ParisTech researchers and makes it freely available over the web where possible. A B S T R A C TNanoporous membranes are used for the elaboration of implantable medical devices. In order to guaranty their integrity after implantation in a patient body, it is necessary to characterize the microstructure and the mechanical behavior of such membranes. They present randomly distributed pores around 1 µm in diameter at the surface. X-ray nanotomography permits to get the geometry of the pores through the thickness with a reduction of the diameter in the core. A multiscale study is done to characterize the membranes: macroscopic tensile tests permit to get the behavior law of the non porous material and in situ tensile tests are carried on in a Scanning Electron Microscope in order to observe the evolution of pores and cracks during loading. A 2D Finite Element Model is also developed in parallel. The confrontation between experiments and numerical simulations permit to validate the accuracy of the model. The latter is then used to simulate several types of loadings considering various pore distributions and sizes.

show abstract

“…From experiments, it was found that when the solid phase holdup reached 4.4%, the particle images overlapped significantly. The overlapped particle count is achieved by detecting the inner and outer edges of the overlapping particle image to find the concave point of the outer contour of the overlapping particle or the center point of the fitting circle, [22][23][24] but the selection of related algorithm parameters is sensitive, and it is easy to be disturbed by noise and cause the misdetection of the contour. Angst et al 25 used a threshold segmentation method to measure the gas phase holdup in a liquid-solid stirring system, but this method has a large error in the extraction of dim bubbles.…”

Section: Introductionmentioning

confidence: 99%

Method to concurrently measure local gas and solid holdups visually with higher precision in circulating fluidized bed

Zhang

et al. 2021

AIChE Journal

View full text Add to dashboard Cite

Simultaneous and visual measurements of local gas and solid phase holdups with high-precision in a gas-liquid-solid circulating fluidized bed is of great importance for the design and scale-up of such a reactor. However, such measuring techniques are still limited due to the difficulty of discrete phase identification of gas bubble and solid particle phases in the complex three-phase flow. Immersed telecentric camera method developed based on the digital image processing is such a measuring technique, but its measuring precision is to be improved due to the existing three problems in the depth-of-field particle extraction, reconstruction of dim bubbles, and counting of overlapping particle clusters. This work has solved these problems by developing corresponding algorithms and the relative errors of measured solid holdup and gas holdup are as low as 0.3% and 1.0%, respectively.

show abstract

Automatic counting method for complex overlapping erythrocytes based on seed prediction in microscopic imaging

Cited by 6 publications

References 21 publications

Wheat Seed Classification: Utilizing Ensemble Machine Learning Approach

Wheat Seed Classification: Utilizing Ensemble Machine Learning Approach

Mechanical properties of a nanoporous membrane used in implantable medical devices. Correlation between experimental characterization and 2D numerical simulation

Method to concurrently measure local gas and solid holdups visually with higher precision in circulating fluidized bed

Contact Info

Product

Resources

About