• Corpus ID: 17513076

An Error Metric for Binary Images

  title={An Error Metric for Binary Images},
  author={Adrian J. Baddeley},
The discrepancy between two binary images is traditionally measured by either thèstatistical' misclassiication error rate, or by Pratt's 1] Figure Of Merit. We discuss weaknesses of these measures, and introduce an improved metric which has both a theoretical basis 2] and intuitive appeal. The error measures are compared on artiicial data and on the standard chessboard test for edge detectors. 

Figures from this paper

Errors in Binary Images and an L P Version of the Hausdorr Metric
A metric is introduced which is an L p modiication of the Hausdorr metric, which generates the myopic topology yet has many features in common with L p metrics, including better robustness against noise.
Quantitative error measures for edge detection
Title Distance-Reciprocal Distortion Measure for Binary Document Images 2 )
  • Haiping Lu, A. Kot
  • Computer Science
  • 2003
The results show that the proposed distortion measure matches well to subjective evaluation by human visual perception.
Distance-reciprocal distortion measure for binary document images
The results show that the proposed distortion measure matches well to subjective evaluation by human visual perception.
A Review on Image Distortion Measures
Within this paper we review image distortion measures. A distortion measure is a criterion that assigns a "quality number" to an image. We distinguish between mathematical distortion measures and
Dissimilarity measures in color spaces
The aim of the paper is to study the behavior of dissimilarity measures in different color spaces, and to compare the results with those obtained by subjective testing.
A New Normalized Supervised Edge Detection Evaluation
In this paper, several referenced-based boundary detection evaluations are detailed, pointing their advantages and disadvantages through concrete examples of edge images, and a new supervised edge map quality measure is proposed, comparing a ground truth contour image, the candidate contours image and their associated spacial nearness.
Towards an objective edge detection algorithm based on discrete t-norms
A comparative analysis of different techniques in order to transform a gray-scale edge image to a thin binary edge image is performed according to some performance measures, such as Pratt's figure of merit, Baddeley's measure and the ρ-coefficient.
Loss Functions for Bayesian Image Analysis
The role of loss functions in Bayesian image classification, object recognition and identification is discussed, and the use of a particular loss function which produces visually attractive estimates is reviewed.
New discrepancy measures for segmentation evaluation
New evaluation measures for scene segmentation results are proposed, which are based on computing the difference between a region extracted from a segmentation map and the corresponding one on an ideal segmentation.


A Quantitative Study of the Orientation Bias of Some Edge Detector Schemes
Improved qualitative observations are reported and a comparison of the evaluation method discussed here with another edge detection evaluation method is presented.
A study of edge detection algorithms
  • T. Peli
  • Computer Science
    Comput. Graph. Image Process.
  • 1982
Distance functions on digital pictures
Comments on "Digital Step Edges from Zero Crossings of Second Directional Derivatives"
Evaluation of the performance of the Marr-Hildreth implementation of the ¿2G operator on similar images shows that this edge detection method in fact performs comparably to the Prewitt and Haralick operators.
On the Statistical Analysis of Dirty Pictures
may 7th, 1986, Professor A. F. M. Smith in the Chair] SUMMARY A continuous two-dimensional region is partitioned into a fine rectangular array of sites or "pixels", each pixel having a particular
A nonlinear laplace operator as edge detector in noisy images
"On the localization performance measure and optimal edge detection"
In this comment, the authors show that although Canny's derivation was in error, the final form of his criterion is adequate and can, in fact, be derived from Tagare and deFigueiredo's formulation of the problem.
Quantitative design and evaluation of enhancement/thresholding edge detectors
Theoretical and experimental comparisons of edge detectors are presented and quantitative design and performance evaluation techniques developed are used to optimally design a variety of small and large mask edge detectors.
Distance transformations in digital images
  • G. Borgefors
  • Computer Science
    Comput. Vis. Graph. Image Process.
  • 1986