Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 27 of the 29 references it cites were found in a registry or an index; none of the 28 sources its claims cite is checked for support yet.

scikit-image: image processing in Python

Open access

Published version of a preprint

Abstract

scikit-image is an image processing library that implements algorithms and utilities for use in research, education and industry applications. It is released under the liberal Modified BSD open source license, provides a well-documented API in the Python programming language, and is developed by an active, international team of collaborators. In this paper we highlight the advantages of open source to achieve the goals of the scikit-image library, and we showcase several real-world image processing applications that use scikit-image. More information can be found on the project homepage, http://scikit-image.org.

Overview

CoChat Summary

TL;DR

scikit-image is an open-source (Modified BSD) Python image processing library built on NumPy arrays, described here alongside its design philosophy, development practices, and real-world research, education, and industry use cases. The paper is a software description rather than an empirical study, presenting the library's architecture, submodules, and example applications.

Key findings

  • The project started in August 2009, has received contributions from more than 100 individuals, and was downloaded more than 5,000 times from PyPI in March 2014 alone.
  • As of version 0.10 the package contains 15 sub-modules (color, data, draw, exposure, feature, filter, graph, io, measure, morphology, novice, restoration, segmentation, transform, viewer).
  • Unit test coverage is "above 87% as of release 0.10 and continues to rise," with continuous integration checking each commit on both Python 2 and Python 3.
  • Images are represented as NumPy arrays under an "Anything In, Anything Out" convention, with floating-point images expected in [0, 1] (unsigned) or [−1, 1] (signed) and 8-bit images in {0, …, 255}.

Methods

  • Design: a descriptive/software paper outlining the library's goals (high-quality documented algorithms, education, industry use), data format, pipelining conventions, and development practices (GitHub pull-request review, PEP8 compliance, Cython for performance-critical sections).
  • Data/sample: illustrative usage examples rather than a study sample — analysis of drying colloidal droplets (crack tracking via Hough transform), zebrafish spinal cord regeneration (measure.profile_line), silicon wafer defect detection at BT Imaging, disguise detection at AICBT, and a Petra panorama stitching demo using ORB features and RANSAC.
  • Main analysis: narrative exposition of API design, licensing, and comparisons with related libraries (Mahotas, ImageJ/Fiji, Matlab's toolbox, OpenCV); no quantitative benchmark or evaluation is reported.

Limitations

  • Stated by the paper: the authors note that their choice of a high-level language "may affect performance," and that strict code review guidelines "may hamper the number of features we ultimately provide"; they also note that Fiji's breadth of functionality is "unparalleled" and that OpenCV offers faster implementations for real-time applications.
  • Evident from the text: full test coverage is listed as a roadmap goal, implying it is not yet achieved; the paper reports no performance benchmarks, accuracy metrics, or user studies to substantiate the claimed advantages.
  • Evident from the text: the usage examples are demonstrations of the library's capabilities rather than controlled evaluations, so no comparative outcome data are provided.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.7717/peerj.453

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Illustration of several functions available in scikit-image: adaptive threshold, local maxima, edge detection and labels. The use of NumPy arrays as our data container also enables the use of NumPy’s built-in histogram function.
Figure 1 Illustration of several functions available in scikit-image: adaptive threshold, local maxima, edge detection and labels. The use of NumPy arrays as our data container also enables the use of NumPy’s built-in histogram function.CC BY · PubMed Central
Figure 2 scikit-image is used to track the propagation of cracks (black lines) in a drying colloidal droplet. The sequence of pictures shows the temporal evolution of the system with the drop contact line, in green, detected by the Hough transform and the circle, in white, used to extract an annulus of pixel intensities. The result shown illustrates the angular position of cracks and their time of appearance.
Figure 2 scikit-image is used to track the propagation of cracks (black lines) in a drying colloidal droplet. The sequence of pictures shows the temporal evolution of the system with the drop contact line, in green, detected by the Hough transform and the circle, in white, used to extract an annulus of pixel intensities. The result shown illustrates the angular position of cracks and their time of appearance.CC BY · PubMed Central
Figure 3 The measure.profile_line function being used to track recovery in spinal cord injuries. (A) An image of fluorescently-labeled nerve cells in an injured zebrafish embryo. (B) The automatically determined region of interest. The SciPy library was used to determine the region extent (Oliphant, 2007; Jones, Oliphant & Peterson, 2001), and functions from the scikit-image draw module were used to draw it. (C) The image intensity along the line of interest, averaged over the displayed width.
Figure 3 The measure.profile_line function being used to track recovery in spinal cord injuries. (A) An image of fluorescently-labeled nerve cells in an injured zebrafish embryo. (B) The automatically determined region of interest. The SciPy library was used to determine the region extent (Oliphant, 2007; Jones, Oliphant & Peterson, 2001), and functions from the scikit-image draw module were used to draw it. (C) The image intensity along the line of interest, averaged over the displayed width.CC BY · PubMed Central
Figure 4 Use of scikit-image to study silicon wafer impurities. (A) An image of an as-cut silicon wafer before it has been processed into a solar cell. (B) A PL image of the same wafer. Wafer defects, which have a negative impact solar cell efficiency, are visible as dark regions. (C) Image processing results. Defects in the crystal growth (dislocations) are colored blue, while red indicates the presence of impurities.
Figure 4 Use of scikit-image to study silicon wafer impurities. (A) An image of an as-cut silicon wafer before it has been processed into a solar cell. (B) A PL image of the same wafer. Wafer defects, which have a negative impact solar cell efficiency, are visible as dark regions. (C) Image processing results. Defects in the crystal growth (dislocations) are colored blue, while red indicates the presence of impurities.CC BY · PubMed Central
Figure 5 An example application of scikit-image: image registration and warping to combine overlapping images. (A) Photographs taken in Petra, Jordan by François Malan. License: CC-BY. (B) Putative matches computed from ORB binary features. (C) Matches filtered using RANSAC. (D) The second input frame (middle) is warped to align with the first input frame (left), yielding the averaged image shown on the right. (E) The final panorama image, registered and warped using scikit-image, blended with Enblend.
Figure 5 An example application of scikit-image: image registration and warping to combine overlapping images. (A) Photographs taken in Petra, Jordan by François Malan. License: CC-BY. (B) Putative matches computed from ORB binary features. (C) Matches filtered using RANSAC. (D) The second input frame (middle) is warped to align with the first input frame (left), yielding the averaged image shown on the right. (E) The final panorama image, registered and warped using scikit-image, blended with Enblend.CC BY · PubMed Central

Verification

Claims the paper makes

0 Valid · 10 Not valid · 21 Not checked · 66 Own results and interpretations

1 more claim

Create a free account to see all 97 claims and to verify the 10 references still open.

See all claims — free account
  • Probably not a substantial claim. 87
  • It may state two separate claims as one. 39
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Show all 29 references

Cited by

Items 7,301–7,400 of 8,191

Metadata

Authors
van der Walt, S., Schönberger, J. L., Nunez-Iglesias, J., Boulogne, F., Warner, J. D., Yager, N., Gouillart, E., & Yu, T.
Affiliations
  • Stellenbosch University (ZA)
  • University of North Carolina at Chapel Hill (US)
  • Melbourne Bioinformatics (AU)
  • Princeton University (US)
  • Mayo Clinic (US)
  • AICBT Ltd, Oxford, UK
  • Centre National de la Recherche Scientifique (FR)
  • Enthought (United States) (US)
Venue
PeerJ
Publisher
PeerJ
Published
2014-06-19
Volume
2
Pages
e453
ISSN
2167-8359 (ISSN-L)
OpenAlex
W2015159529
Cited by
6,794
References
29
Citations by year
2026: 846 · 2025: 1046 · 2024: 914 · 2023: 915 · 2022: 867 · 2021: 768 · 2020: 507 · 2019: 366 · 2018: 227 · 2017: 165 · 2016: 102 · 2015: 50 · 2014: 15
FWCI
110.33
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • PeerJ (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • Minerva Access (University of Melbourne) (submittedVersion, cc-by)
  • HAL (submittedVersion, other-oa)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • SUNScholar (Stellenbosch University) (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Funders
National Institute of Diabetes and Digestive and Kidney Diseases (F30DK098832)
Topics
  • Physical Sciences › Computer Science › Software › Computational Physics and Python Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › Computer Science and Engineering
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Image Processing Techniques
Keywords
scikit-image, image processing, Python library, open source software, image processing algorithms, real-world applications

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-18, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-18 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.