Geometric precision in noise-free digital images

Havelock, David I.

doi:10.1109/34.42837

Cited by 45 publications

(19 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…due to repeated scanning was exploited by Zhou and Lopresti to decrease OCR error [20]. Random-phase sampling noise was systematically investigated in remote sensing [21,22] and in scanned documents [23], and eventually incorporated into Henry Baird"s defect model [24]. The relationship between spatial and amplitude quantization in scanning was explored thoroughly by Barney Smith [25].…”

Section: Prior Workmentioning

confidence: 99%

On Parallel Lines in Noisy Forms

Nagy

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Quantification of the rectilinear configuration of typeset rules (lines) opens the way to form classification and content extraction. Line detection on scanned forms is often accomplished with the Hough transform. Here it is followed by simultaneous extraction of the dominant perpendicular sets of extracted lines, which ensures rotation invariance. Translation and scale invariance are attained by using minimal horizontal and vertical sets of distance ratios ("rule gap ratios") instead of rule-edge locations. The ratios are logarithmically mapped to an alphabet so that the resulting symbol strings can be classified by. edit distance. Some probability distributions associated with these steps are derived. Analytical considerations and small-scale experiments on scanned forms suggest that this approach has potential merit for highly degraded forms.

show abstract

Section: Prior Workmentioning

confidence: 99%

On Parallel Lines in Noisy Forms

Nagy

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Recently, some evidence of the relative superiority of the centroid method for signal localization over such methods as zero crossing or peak detection has come to light [4]. In practice, the centroid method not only resulted in a substantial improvement in resolvable detail, it was remarkably robust in the presence of noise, low signal level, and nonlinearities such as those caused by surface features.…”

Section: Marketing As the Mother Of Inventionmentioning

confidence: 99%

Technology innovations and product design issues in machine vision: The Technical Arts Corporation experience

White¹

1990

Procedings of the British Machine Vision Conference 1990

View full text Add to dashboard Cite

“…In [6] the loss of accuracy in digitized straight lines coming from quantization is analyzed giving a quantitative idea of damages coming from one of the steps implied from digitalization. The work reported in [7] introduced the idea of Locales, which are broadly applicable to the estimation of the position and shape of objects, although their use is showed only for binary images. In [8] an interesting mathematical tool for the estimation of the average error due to quantization is introduced, but it does not take into account a sufficiently detailed number of factors to be useful for a practical application in evaluating the uncertainty of images.…”

Section: Introductionmentioning

confidence: 99%