On Detection of Faint Edges in Noisy Images

Nati Ofir; Meirav Galun; Sharon Alpert; Achi Brandt; Boaz Nadler; Ronen Basri

doi:10.1109/TPAMI.2019.2892134

On Detection of Faint Edges in Noisy Images

Nati Ofir, Meirav Galun, Sharon Alpert, Achi Brandt, Boaz Nadler, Ronen Basri

Weizmann Institute of Science

Research output: Contribution to journal › Article › peer-review

31 Scopus citations

Abstract

A fundamental question for edge detection in noisy images is how faint can an edge be and still be detected. In this paper we offer a formalism to study this question and subsequently introduce computationally efficient multiscale edge detection algorithms designed to detect faint edges in noisy images. In our formalism we view edge detection as a search in a discrete, though potentially large, set of feasible curves. First, we derive approximate expressions for the detection threshold as a function of curve length and the complexity of the search space. We then present two edge detection algorithms, one for straight edges, and the second for curved ones. Both algorithms efficiently search for edges in a large set of candidates by hierarchically constructing difference filters that match the curves traced by the sought edges. We demonstrate the utility of our algorithms in both simulations and applications involving challenging real images. Finally, based on these principles, we develop an algorithm for fiber detection and enhancement. We exemplify its utility to reveal and enhance nerve axons in light microscopy images.

Original language	English
Article number	8607091
Pages (from-to)	894-908
Number of pages	15
Journal	IEEE Transactions on Pattern Analysis and Machine Intelligence
Volume	42
Issue number	4
DOIs	https://doi.org/10.1109/TPAMI.2019.2892134
State	Published - 1 Apr 2020
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 1979-2012 IEEE.

Funding

Research was supported in part by the Institute for Future Defense Technologies Research named for the Medvedi, Shwartzman and Gensler Families, and by the European Commission Project IST-2002-506766 Aim Shape. Part of this research was conducted while RB was at TTI-C. At the Weizmann Institute research was conducted at the Moross Laboratory for Vision and Motor Control. We thank Eyal Shimoni and Ziv Reich for the images in Fig. 1 and Ida Rishal and Mike Fainzilber for the images in Fig. 14. We thank Pedro Felzenswalb for sharing his insights with us.

Funders	Funder number
Institute for Future Defense Technologies Research named for the Medvedi, Shwartzman and Gensler Families
European Commission	IST-2002-506766

Keywords

Edge detection
fiber enhancement
low signal-to-noise ratio
microscopy images
multiple hypothesis tests
multiscale methods

Access to Document

10.1109/TPAMI.2019.2892134

Cite this

@article{6d1ef2bed3e444a38d69edbd28f565e7,

title = "On Detection of Faint Edges in Noisy Images",

abstract = "A fundamental question for edge detection in noisy images is how faint can an edge be and still be detected. In this paper we offer a formalism to study this question and subsequently introduce computationally efficient multiscale edge detection algorithms designed to detect faint edges in noisy images. In our formalism we view edge detection as a search in a discrete, though potentially large, set of feasible curves. First, we derive approximate expressions for the detection threshold as a function of curve length and the complexity of the search space. We then present two edge detection algorithms, one for straight edges, and the second for curved ones. Both algorithms efficiently search for edges in a large set of candidates by hierarchically constructing difference filters that match the curves traced by the sought edges. We demonstrate the utility of our algorithms in both simulations and applications involving challenging real images. Finally, based on these principles, we develop an algorithm for fiber detection and enhancement. We exemplify its utility to reveal and enhance nerve axons in light microscopy images.",

keywords = "Edge detection, fiber enhancement, low signal-to-noise ratio, microscopy images, multiple hypothesis tests, multiscale methods",

author = "Nati Ofir and Meirav Galun and Sharon Alpert and Achi Brandt and Boaz Nadler and Ronen Basri",

note = "Publisher Copyright: {\textcopyright} 1979-2012 IEEE.",

year = "2020",

month = apr,

day = "1",

doi = "10.1109/TPAMI.2019.2892134",

language = "אנגלית",

volume = "42",

pages = "894--908",

journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",

issn = "0162-8828",

publisher = "IEEE Computer Society",

number = "4",

}

TY - JOUR

T1 - On Detection of Faint Edges in Noisy Images

AU - Ofir, Nati

AU - Galun, Meirav

AU - Alpert, Sharon

AU - Brandt, Achi

AU - Nadler, Boaz

AU - Basri, Ronen

PY - 2020/4/1

Y1 - 2020/4/1

N2 - A fundamental question for edge detection in noisy images is how faint can an edge be and still be detected. In this paper we offer a formalism to study this question and subsequently introduce computationally efficient multiscale edge detection algorithms designed to detect faint edges in noisy images. In our formalism we view edge detection as a search in a discrete, though potentially large, set of feasible curves. First, we derive approximate expressions for the detection threshold as a function of curve length and the complexity of the search space. We then present two edge detection algorithms, one for straight edges, and the second for curved ones. Both algorithms efficiently search for edges in a large set of candidates by hierarchically constructing difference filters that match the curves traced by the sought edges. We demonstrate the utility of our algorithms in both simulations and applications involving challenging real images. Finally, based on these principles, we develop an algorithm for fiber detection and enhancement. We exemplify its utility to reveal and enhance nerve axons in light microscopy images.

AB - A fundamental question for edge detection in noisy images is how faint can an edge be and still be detected. In this paper we offer a formalism to study this question and subsequently introduce computationally efficient multiscale edge detection algorithms designed to detect faint edges in noisy images. In our formalism we view edge detection as a search in a discrete, though potentially large, set of feasible curves. First, we derive approximate expressions for the detection threshold as a function of curve length and the complexity of the search space. We then present two edge detection algorithms, one for straight edges, and the second for curved ones. Both algorithms efficiently search for edges in a large set of candidates by hierarchically constructing difference filters that match the curves traced by the sought edges. We demonstrate the utility of our algorithms in both simulations and applications involving challenging real images. Finally, based on these principles, we develop an algorithm for fiber detection and enhancement. We exemplify its utility to reveal and enhance nerve axons in light microscopy images.

KW - Edge detection

KW - fiber enhancement

KW - low signal-to-noise ratio

KW - microscopy images

KW - multiple hypothesis tests

KW - multiscale methods

UR - http://www.scopus.com/inward/record.url?scp=85073893421&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2019.2892134

DO - 10.1109/TPAMI.2019.2892134

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

C2 - 30629496

AN - SCOPUS:85073893421

SN - 0162-8828

VL - 42

SP - 894

EP - 908

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

IS - 4

M1 - 8607091

ER -

On Detection of Faint Edges in Noisy Images

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this