Elsevier

Pattern Recognition

Volume 29, Issue 5, May 1996, Pages 829-843
Pattern Recognition

Digital image thresholding, based on topological stable-state

https://doi.org/10.1016/0031-3203(95)00126-3Get rights and content

Abstract

A new approach for image segmentation for scenes that contain distinct objects is presented. A sequence of graphs Ns(t) is defined, where Ns(t) is the number of connected objects composed of at least s pixels, for the image thresholded at t. The sequence of graphs is built in almost linear time complexity, namely at O(α(n, nn), where α(n, n) is the inverse of the Ackermann function, and n is the number of pixels in the image. Stable states on the graph in the appropriate “resolution” s* correspond to threshold values that yield a segmentation similar to a human observer. The relevance of a Percolation model to the graphs Ns(t) is discussed.

References (26)

  • PalN.R. et al.

    Image thresholding: some new techniques

    Signal Process.

    (1993)
  • NakagawaY. et al.

    Some experiments on variable thresholding

    Pattern Recognition

    (1979)
  • ChowC.K. et al.

    Automatic boundary detection of the left ventricle from cineangiongrams

    Comput. Biomed. Res.

    (1972)
  • Cited by (68)

    • A review of remote sensing applications for water security: Quantity, quality, and extremes

      2020, Journal of Hydrology
      Citation Excerpt :

      entropy-based methods – determine the entropy of the two clusters and the cross-entropy between the original and binarized images (Cheng et al., 1999; Kapur et al., 1985; Rajinikanth et al., 2018; Sahoo et al., 1997; Sarkar et al., 2016), object attribute-based methods – which assess the similarity between original and binarized images (Dasgupta et al., 2018; Hertz and Schafer, 1988; Martinis et al., 2009; Pikaz and Averbuch, 1996; Pulvirenti et al., 2011), spatial methods – which use probability distributions between the pixels (Brivio et al., 2002; Cao et al., a,b,c).

    • Imaged based fractal characterization of micro-fracture structure in coal

      2019, Fuel
      Citation Excerpt :

      After the coal images treated with denoise, the gray images were segmented, this was accomplished by setting a threshold value first, and then the grayscale of the pixels which were bigger than the threshold were set as 255 or set up as 0 if the grayscale of the pixels were smaller than the threshold. This process is termed as binarization, which is essential before pore and/or fracture extraction from rock images [37,38]. Several methods have been used to determine the threshold for binarization, such as the bimodal method, porosity restriction method and the Digital Terrain Model (DTM).

    • Different binarization processes validated against manual counts of fluorescent bacterial cells

      2016, Journal of Microbiological Methods
      Citation Excerpt :

      In this method, groups of objects with identical intensity-values are predefined and correlated to an incremental increasing threshold level. Objects regarded as “noise”, rapidly disappear when shifting the threshold (Pikaz and Averbuch, 1996). Another comparable algorithm is the stable count method.

    • Optimal multi-level thresholding with membrane computing

      2015, Digital Signal Processing: A Review Journal
    • Segmentation of color lip images by optimal thresholding using bacterial foraging optimization (BFO)

      2014, Journal of Computational Science
      Citation Excerpt :

      Indeed, image thresholding is a major operation in many image processing applications such as optical character recognition where the goal is to extract the character in a document image and then recognize it [4]. Thresholding techniques can be classified into two categories: The first category includes methods that find the optimal threshold using image histogram analysis [5]. The second category includes methods that find the optimal threshold using objective functions.

    View all citing articles on Scopus
    View full text