2D Otsu Segmentation Algorithm Improvement Based on FOCPSO

Zhang, Yingjie; Zeng, Liang; Zhang, Ying; Meng, Jie

doi:10.1109/bdcloud.2018.00121

Cited by 2 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the modifications approach, there are some papers which suggest acceleration of the Otsu criterion calculation based on the genetic optimization algorithms [37] and GPU-accelerated computations [38].…”

Section: An Overview Of the Known Otsu Methods Modificationsmentioning

confidence: 99%

A generalization of Otsu method for linear separation of two unbalanced classes in document image binarization

et al. 2021

View full text Add to dashboard Cite

The classical Otsu method is a common tool in document image binarization. Often, two classes, text and background, are imbalanced, which means that the assumption of the classical Otsu method is not met. In this work, we considered the imbalanced pixel classes of background and text: weights of two classes are different, but variances are the same. We experimentally demonstrated that the employment of a criterion that takes into account the imbalance of the classes' weights, allows attaining higher binarization accuracy. We described the generalization of the criteria for a two-parametric model, for which an algorithm for the optimal linear separation search via fast linear clustering was proposed. We also demonstrated that the two-parametric model with the proposed separation allows increasing the image binarization accuracy for the documents with a complex background or spots.

show abstract

Section: An Overview Of the Known Otsu Methods Modificationsmentioning

confidence: 99%

A generalization of Otsu method for linear separation of two unbalanced classes in document image binarization

et al. 2021

View full text Add to dashboard Cite

show abstract

“…However, the method assumes that the influence of edge region information on image segmentation can be ignored. This assumption makes the classical 2D Otsu method ineffective for segmenting images with rich edge information [18]. Therefore, subsequent researchers have given many improved methods: Liang et al [19] introduced the idea of iterative segmentation based on the original 2D Otsu thresholding segmentation method, after obtaining the maximum inter-class variance threshold point for the whole image grayscale, the edge region continues to be segmented iteratively so that a series of threshold points are obtained, and a fold line threshold can be obtained by connecting each threshold point in turn, the advantages of this method are that line threshold replaces traditional point threshold and improves the accuracy of the segmentation, while the disadvantages are that there is region wrong classification and the number of iterations is artificially given based on experience, which reduces the adaptiveness.…”

Section: Introductionmentioning

confidence: 99%

Two-dimensional Otsu’s Zigzag Thresholding Segmentation Method

Liang

chen

2023

Preprint

View full text Add to dashboard Cite

In the two-dimensional histogram thresholding method, the segmentation accuracy could be degraded due to insufficient misclassification. An improved image segmentation method combining a priori knowledge of two-dimensional histogram and fully considering the influence of edges, the two-dimensional Otsu’s zigzag thresholding segmentation method is proposed. We combine a priori information about the edge regions and noisy regions in a two-dimensional histogram, use a zigzag threshold as a segmentation criterion to correct the overall error classification and use small probability events to determine the line equations to achieve segmentation adaptively. Based on extensive experimentation, our method has been observed to significantly outperform comparable techniques.

show abstract

2D Otsu Segmentation Algorithm Improvement Based on FOCPSO

Cited by 2 publications

References 8 publications

A generalization of Otsu method for linear separation of two unbalanced classes in document image binarization

A generalization of Otsu method for linear separation of two unbalanced classes in document image binarization

Two-dimensional Otsu’s Zigzag Thresholding Segmentation Method

Contact Info

Product

Resources

About