Multilevel thresholding of images with improved Otsu thresholding by black widow optimization algorithm

ALRAHLAWEE>, Anfal Thaer; Rahebi, Javad

doi:10.1007/s11042-021-10860-w

Cited by 41 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Equations ( 22 ), ( 23 ), ( 24 ), and ( 25 ) show how to calculate accuracy, sensitivity, precision, and F1, respectively (Al-Rahlawee and Javad Rahebi. 2021 ; Albargathe et al 2021 ; Alhayani and Abdallah 2020 ; Alhayani and Llhan 2021 ): …”

Section: Methodsmentioning

confidence: 99%

RETRACTED ARTICLE: Feature selection for diagnose coronavirus (COVID-19) disease by neural network and Caledonian crow learning algorithm

alsaeedi

Kurnaz

2022

Appl Nanosci

View full text Add to dashboard Cite

In this study, feature selection methods based on the new Caledonian crow learning algorithm has been introduced. In the proposed algorithms, in the first stage, the best features related to COVID-19 disease are selected by the crow learning algorithm. Coronavirus (COVIDE-19) disease using as training input to the artificial neural network. Experiments on the COVID-19 disease dataset in a Brazilian hospital show that the crow learning algorithm reduces the feature selection objective function by iteration. Decreasing the feature selection function is due to reducing the error of classifying infected people as healthy and reducing the number of features. The experimental results show that the accuracy, sensitivity, precision, and F1 of the proposed method for COVID-19 patients diagnosing are 94.31%, 94.15%, 94.38%, and 94.27%, respectively. The proposed method for identifying COVID-19 patients is more accurate than ANN, CNN, CNNLSTM, CNNRNN, LSTM, and RNN methods.

show abstract

Section: Methodsmentioning

confidence: 99%

RETRACTED ARTICLE: Feature selection for diagnose coronavirus (COVID-19) disease by neural network and Caledonian crow learning algorithm

alsaeedi

Kurnaz

2022

Appl Nanosci

View full text Add to dashboard Cite

show abstract

“…Orthomosaic maps were generated using Agisoft Metashape Pro Version 1.7.2 (2021); these maps were draped over Digital Terrain Models (DTMs) to document melt patterns using structure‐from‐motion (SfM) photogrammetric techniques (Nolan et al., 2015). We generated binary images of snow/no‐snow areas, the basis for computing snow cover fraction, using the Otsu method (Al‐Rahlawee & Rahebi, 2021).…”

Section: Measurements and Methodsmentioning

confidence: 99%

An Examination of Water‐Related Melt Processes in Arctic Snow on Tundra and Sea‐Ice

Pinzner,

Sturm,

Delamere

et al. 2024

Water Resources Research

View full text Add to dashboard Cite

From April through June in 2019 and 2022, we monitored snow melt at three sites near Utqiaġvik, Alaska. Along 200‐m lines we measured snow depth, density, stratigraphy, snow‐covered area, and spectral albedo. Site 1 (ARM) was sloped tundra drained by water tracks. Site 2 (BEO) was flat polygonal tundra. Site 3 (ICE) was on undeformed landfast sea ice. All three sites were within a 6 km radius. Despite similar pre‐melt snow distributions and weather, the melt progression differed markedly between sites. In 2019, by mid‐melt, there was 40% less snow‐covered area at ARM versus ICE, and 34% less snow‐covered area at ARM versus BEO. The 2022 melt started 2 weeks later than in 2019 and was rapid, so smaller differences in snow‐covered areas developed. In both years meltout dates varied by up to 25 days between sites, and more than 20 days within sites, with melt rates at locations only meters apart differing by up to a factor of seven. This melt diachroneity led to highly heterogeneous meltout patterns at all three sites. Our measurements and observations indicate that, in addition to reductions in snow reflective properties and wind‐driven heat advection, the fate of meltwater plays a key role in producing melt diachroneity. We identify seven snow‐water mechanisms that can enhance or inhibit melt rates, all largely controlled by the local topography and the nature of the substrate. These mechanisms are important because the most rapid changes in albedo coincide with the peak of water‐snow melt interactions.

show abstract

“…The MobileNet‐based cloud contamination removal and enhancement algorithm was proposed for cloud detection and extraction. Compressed signal sensing in the time and frequency domain led to image improvement 28 . Finally, a sparse representation model generated from the hyperspectral image relied on the compressed image for analysis.…”

Section: Related Workmentioning

confidence: 99%

“…Compressed signal sensing in the time and frequency domain led to image improvement. 28 Finally, a sparse representation model generated from the hyperspectral image relied on the compressed image for analysis. Successful image enhancement algorithms were determined by an area overlap measure (AOM) of 0.83 and an average gradient of 12.7.…”

Section: Feature Optimization Algorithmsmentioning

confidence: 99%

Improved multiple sclerosis diagnosis with advanced deep learning techniques

Pandian¹,

Udhayakumar²

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

White matter pathologies analysis is important for segmenting multiple sclerosis (MS) lesions in MRI slices. Myelin damage affects motor function, sensory conflicts, paralysis, and vision. Extracting an optimistic number of hidden characteristic features induces high feature maps and reduces redundancy and dimensionality. A framework based on Hybrid Deep Convolutional Neural Networks (HD‐CNN) is presented to classify MS based on the acquired optimistic feature characteristic. First, Chaotic Leader‐Selective Particle Swarm Optimization uses the pixel intensity ratio of the targeted region to extract the most hidden white matter features. Second, Refined Slime Mold categorizes hidden features and ensures that the necessary features are selected effectively for a better segmentation result. An effective inbuilt Maximum Bidirectional Gradient Classifier accurately classifies MS lesions by detecting white matter spots in the target region. Finally, we will evaluate the proposed HD‐CNN framework using the industry‐standard University Medical Centre Ljubljana (UMCL) and international symposium on biomedical imaging (ISBI) 2015 datasets as benchmarks. It reaches 98.56%, 93.15%, and 98.44% accuracy for FLAIR, T1W, and T2W, respectively. Our HD‐CNN improves MS classification over other state‐of‐the‐art techniques, compared with accuracy, F‐score, Recall, Precision, Dice, and Jaccard indexes values.

show abstract

Multilevel thresholding of images with improved Otsu thresholding by black widow optimization algorithm

Cited by 41 publications

References 28 publications

RETRACTED ARTICLE: Feature selection for diagnose coronavirus (COVID-19) disease by neural network and Caledonian crow learning algorithm

RETRACTED ARTICLE: Feature selection for diagnose coronavirus (COVID-19) disease by neural network and Caledonian crow learning algorithm

An Examination of Water‐Related Melt Processes in Arctic Snow on Tundra and Sea‐Ice

Improved multiple sclerosis diagnosis with advanced deep learning techniques

Contact Info

Product

Resources

About