A threshold selection method from gray level histograms

  title={A threshold selection method from gray level histograms},
  author={Nobuyuki Otsu},
  journal={IEEE Transactions on Systems, Man, and Cybernetics},
  • N. Otsu
  • Published 1979
  • Engineering
  • IEEE Transactions on Systems, Man, and Cybernetics
A nonparametric and unsupervised method ofautomatic threshold selection for picture segmentation is presented. An optimal threshold is selected by the discriminant criterion, namely, so as to maximize the separability of the resultant classes in gray levels. The procedure is very simple, utilizing only the zerothand the first-order cumulative moments of the gray-level histogram. It is straightforward to extend the method to multithreshold problems. Several experimental results are also… 

Figures from this paper

An improved threshold selection method for image segmentation

  • X. YangV. Gupta
  • Computer Science
    Proceedings of Canadian Conference on Electrical and Computer Engineering
  • 1993
An unsupervised optimal multi-threshold selection scheme for image segmentation is presented. This method is clearly an improvement on two existing methods, namely, Otsu's (1979) optimal

Threshold Selection Method Based on the Distribution of Gray Levels

This paper presents a simple but effective example showing that the histogram-based thresholding methods do not perform well, and proposes a new gray level threshold selection method based on the distribution of gray levels in images.

Segmentation of infrared image using adaptive thresholding

The theoretical analysis shows that correct image segmentation can be obtained for a low-contrast image with complex background using the proposed novel adaptive method of automatic threshold selection, no matter the proportion of the target is small or big.

Automatic threshold selection based on histogram modes and a discriminant criterion

A fully automatic thresholding method via histogram modal decomposition by data-dependent-systems methodology is presented, in which the histogram of an image is parametrically modeled by the power spectrum of an autoregressive model to provide vital information about histogram clusters.

A threshold selection method based on edge preserving

A method of automatic threshold selection for image segmentation by maximizing the edge energy function is presented and the shortcoming of Otsu's method based on gray-level histograms is analyzed.

Minimum Error Thresholding Based on Two Dimensional Histogram

The purpose of this paper is to extend the minimum error thresholding to the 2dimensional histogram to come up a new algorithm, namely, minimum errorresholding based on the 2D histogram, which can derive the 1D optimal threshold vector by presuming that the object and background gray level values are normally distributed in the 2 D histogram.

Statistical thresholding method for infrared images

The proposed novel thresholding method was compared with several classic thresholding counterparts on a variety of infrared images as well as general real-world ones, and the experimental results demonstrate its superiority.



A Threshold Selection Technique

A method of handling cases in which the peaks are very unequal in size and the valley is broad is described, in which points that lie on or near the edges of objects are determined.

Automatic boundary detection of the left ventricle from cineangiograms.

  • C. ChowT. Kaneko
  • Physics
    Computers and biomedical research, an international journal
  • 1972

A branching algorithm for discriminating and tracking multiple objects

A recursive branching algorithm for multiple-object discrimination and tracking consists of a bank of parallel filters of the Kalman form, each of which estimates a trajectory associated with a

An Optimal Data Association Problem in Surveillance Theory

  • R. Sittler
  • Mathematics
    IEEE Transactions on Military Electronics
  • 1964
A method of inverse probability is obtained which permits maximum likelihood estimates to be made of the true data-surveillance object association and the maximum likelihood estimator is given in a form that lends itself to sequential computations performed in real time as the data arrives.

Finding Events in a Sea of Bubbles

DAPR is a digital automatic pattern recognition system, designed to find and measure events in bubble chamber film without manual intervention. It is able to measure film which has come directly from


A system with three basic components: 1) a sensor which rigorously and effectively scans microscopic fields and converts the optical information into digital form; 2) human analysts who contribute heuristics, devise image processing methods and 3) quantitative methods which fully utilize measurements of light intensity of optical density.

An automated apparatus for cancer prescreening: CYBEST

Book reviews - Orthogonal transforms for digital signal processing

  • G. Redinbo
  • Physics
    IEEE Communications Society Magazine
  • 1977
The combination of topical content and the clarity of explanation in this book when mixed with proven technical writing successes of the authors over the last 30 years assures this reviewer that this volume will become a well-recognized text for practical engineering in digital integrated electronics.

Reduction of Sea Surveillance Data Using Binary Matrices

  • Patrick L. Smith
  • Computer Science
    IEEE Transactions on Systems, Man, and Cybernetics
  • 1976
A new approach is described for assembling feasible ship trajectories from ship position sightings using logical AND and OR operations, analogous to the Kalman filter.

Application of 0-1 integer programming to a track assembly problem

  • C. Morefield
  • Computer Science
    1975 IEEE Conference on Decision and Control including the 14th Symposium on Adaptive Processes
  • 1975
This paper discusses a formulation of the tracking problem based on 0-1 integer programming which can markedly reduce Computer time requirements.