pith. sign in

arxiv: 2604.24471 · v1 · submitted 2026-04-27 · 🌌 astro-ph.GA

Machine learning technique for morphological classification of galaxies from SDSS. IV. Visual inspection vs CNN for merging, irregular, edge-on, barred, ringed, and with dust lanes galaxies at 0.02<z<0.1

Dobrycheva D.V. , Vavilova I.B. , Kompaniiets O.V. , Khramtsov V. , Vasylenko M.Yu. , Hetmantsev O.O. , Melnyk O.V. , Karachentseva V.E This is my paper

Pith reviewed 2026-05-08 02:26 UTC · model grok-4.3

classification 🌌 astro-ph.GA
keywords galaxiesbarrededge-onirregularmorphologicalringeddust-lanegalaxy
0
0 comments X

The pith

Visual inspection of CNN outputs from SDSS produces verified catalogues of 612 merging, 9372 irregular, 16822 edge-on, 575 dust-lane, 811 barred and 2150 ringed galaxies at 0.02<z<0.1 together with BPT-based nuclear activity types.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Astronomers have millions of galaxy images from surveys like SDSS. A computer program called a convolutional neural network was trained to spot six specific shapes: galaxies that look like they are merging, look irregular, appear edge-on, have bars, have rings, or show dust lanes. The program flagged thousands of candidates. Because computers still make mistakes on faint features or projection effects, the team looked at every flagged galaxy by eye. They kept only the ones that really matched the class, removed contaminants such as foreground stars, and produced six clean lists. They also used spectra to classify the central activity of the galaxies as star-forming, LINER-like or composite. Five new polar-ring candidates were noted. The work shows where the computer fails most often and supplies ready-to-use samples for studies of galaxy structure and evolution.

Core claim

We present catalogues of 612 merging, 9,372 irregular, 16,822 edge-on, 575 dust-lane, 811 barred, and 2,150 ringed galaxies. CNN misclassifications stem primarily from projection effects, foreground stars, faint tidal features, and irregular star-forming structures.

Load-bearing premise

That visual inspection by the team provides an unbiased and complete ground truth for the six morphological classes, with no systematic differences in how different inspectors classified ambiguous cases.

Figures

Figures reproduced from arXiv: 2604.24471 by Dobrycheva D.V., Hetmantsev O.O., Karachentseva V.E, Khramtsov V., Kompaniiets O.V., Melnyk O.V., Vasylenko M.Yu., Vavilova I.B..

Figure 1
Figure 1. Figure 1: Distribution of apparent magnitude in r-band of SDSS after visual inspection for: merging galaxies (left); irregular galaxies view at source ↗
Figure 2
Figure 2. Figure 2: Galaxies marked by CNN as merging, but after visual view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of apparent magnitude in r-band of SDSS after visual inspection for: dust lane galaxies (left); bar galaxies view at source ↗
Figure 5
Figure 5. Figure 5: Polar ring galaxy, which is discovered among galaxies view at source ↗
Figure 4
Figure 4. Figure 4: Galaxies with dust lane the dust lanes depends on the galaxy’s stellar mass. It appears at M∗ ∼ 109M⊙ and does not depend on the thickness of the disc or the Sérsic profile, but it correlates with the morphology of the bulge. The central component along the line of sight that pro￾duces the dust lane is not associated with the location of either the cold diffuse component or the heated component in the H II… view at source ↗
Figure 7
Figure 7. Figure 7: Galaxies with ring We performed a cross-matching of galaxies with ring with the Uniformly Selected, All-sky Optical AGN Catalogue for red￾shifts z < 0.09 (Zaw et al. 2019), which is based on optical spectroscopy and includes 1,929 broad-line AGNs and 6,562 narrow-line AGNs. In total, we identified 203 ringed galaxies with AGNs. Unlike inactive galaxies with ring, the number of AGN-hosting galaxies drops sh… view at source ↗
Figure 8
Figure 8. Figure 8: The BPT diagram for galaxies with ring based on SDSS spectra This distribution indicates that most galaxies with rings in our catalog are dominated by low-ionization nuclear emission line (LINER) regions, suggesting either a predominance of weak AGN activity or the presence of shocks. Rings in disc galaxies are frequently associated with non￾axisymmetric structures and secular evolution, where bars and evo… view at source ↗
Figure 9
Figure 9. Figure 9: The BPT diagram for galaxies with bar based view at source ↗
Figure 10
Figure 10. Figure 10: The BPT diagram for galaxies with dust lane based on SDSS spectra unbarred systems. Recent work employing more explicit sample￾matching strategies (e.g., propensity score matching) revisited this question and found evidence for a statistically significant increase in AGN fuelling/incidence in barred galaxies after con￾trolling for key covariates (Silva-Lima et al. 2022). Moreover, using a volume-limited S… view at source ↗
read the original abstract

Context. Convolutional neural networks (CNNs) are widely used for automated galaxy morphological classification in large surveys. However, projection effects, image artefacts, and intrinsic degeneracies limit reliable identification of detailed features, requiring large-scale visual validation. Aims. To visually inspect SDSS galaxies at 0.02 < z < 0.1 classified by a CNN as merging, irregular, edge-on, barred, ringed, or dust-lane galaxies; assess CNN completeness and failure modes; construct visually verified morphological catalogues; and determine nuclear activity types via BPT diagrams. Methods. We visually inspected all galaxies assigned by the CNN to six morphological classes: merging (2,574), irregular (9,432), edge-on (17,000), barred (6,000), ringed (13,882), and dust-lane (588), regardless of CNN probability. Refined samples were cross-matched with Galaxy Zoo 2; remaining galaxies were classified here for the first time. Nuclear activity was determined from SDSS DR17 spectra using H{\alpha}\alpha {\alpha}, H\b{eta}\beta \b{eta}, [O III]\lambda$5007, and [N II] \lambda$6583 line ratios. Results. We present catalogues of 612 merging, 9,372 irregular, 16,822 edge-on, 575 dust-lane, 811 barred, and 2,150 ringed galaxies. CNN misclassifications stem primarily from projection effects, foreground stars, faint tidal features, and irregular star-forming structures. We characterise nuclear activity types for edge-on, barred, ringed, and dust-lane galaxies, finding systematic differences in LINER-like and composite fractions across subsamples. Five strong polar ring galaxy candidates were identified. Conclusions. Visual validation remains essential for refining CNN-based classifications. The resulting datasets support morphological studies, investigations of galaxy structure and secular evolution, and provide robust training samples for future machine learning models.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

No physical model or derivation is introduced; the work rests on standard SDSS data reduction, the pre-existing CNN classifier from earlier papers in the series, and expert visual classification whose consistency is not quantified in the abstract.

axioms (1)
  • domain assumption Visual classification by the listed authors provides a reliable ground truth for the six morphological classes
    Invoked implicitly when the authors state that refined samples were obtained by visual inspection

pith-pipeline@v0.9.0 · 5736 in / 1397 out tokens · 57327 ms · 2026-05-08T02:26:31.083348+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

57 extracted references · 57 canonical work pages

  1. [1]

    K., et al

    Abraham, L., Abraham, S., Kembhavi, A. K., et al. 2025, ApJ, 978, 137 Aguilar-Argüello, G., Fuentes-Pineda, G., Hernández-Toledo, H. M., et al. 2025, MNRAS, 537, 876

    work page 2025

  2. [2]

    D., Sormani, M

    Anderson, L. D., Sormani, M. C., Ginsburg, A., et al. 2020, ApJ, 901, 51 Argudo-Fernández, M., Gómez Hernández, C., Verley, S., et al. 2024, A&A, 692, A258

    work page 2020

  3. [3]

    Bagley, M., Minchev, I., & Quillen, A. C. 2009, Monthly Notices of the Royal Astronomical Society, 395, 537

    work page 2009

  4. [4]

    2011, VizieR Online Data Catalog: EFIGI catalogue of 4458 nearby galaxies (de Lapparent+, 2011), VizieR On-line Data Catalog: J/A+A/532/A74

    Baillard, A., Bertin, W., de Lapparent, V ., et al. 2011, VizieR Online Data Catalog: EFIGI catalogue of 4458 nearby galaxies (de Lapparent+, 2011), VizieR On-line Data Catalog: J/A+A/532/A74. Originally published in: 2011A&A...532A..74B

    work page 2011

  5. [5]

    A., Phillips, M

    Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, Publications of the Astro- nomical Society of the Pacific, 93, 5

    work page 1981

  6. [6]

    J., et al

    Banerji, M., Lahav, O., Lintott, C. J., et al. 2010, MNRAS, 406, 342

    work page 2010

  7. [7]

    2016, Monthly Notices of the Royal Astronomical Society, 461, 3111

    Belfiore, F., Maiolino, R., Maraston, C., & et al. 2016, Monthly Notices of the Royal Astronomical Society, 461, 3111

    work page 2016

  8. [8]

    W., Wilkinson, S., Ferreira, L., et al

    Bickley, R. W., Wilkinson, S., Ferreira, L., et al. 2024, MNRAS, 534, 2533

    work page 2024

  9. [9]

    J., Sheth, K., Athanassoula, E., et al

    Buta, R. J., Sheth, K., Athanassoula, E., et al. 2015, ApJS, 217, 32

    work page 2015

  10. [10]

    & Morgan, L

    Butterfield, N. & Morgan, L. 2024, in American Astronomical Society Meet- ing Abstracts, V ol. 243, American Astronomical Society Meeting Abstracts, 262.19

    work page 2024

  11. [11]

    K., Bekki, K., & Groves, B

    Cavanagh, M. K., Bekki, K., & Groves, B. A. 2021, MNRAS, 506, 659

    work page 2021

  12. [12]

    J., Aragón-Salamanca, A., et al

    Cheng, T.-Y ., Conselice, C. J., Aragón-Salamanca, A., et al. 2020, MNRAS, 493, 4209 Chrobáková, Ž., Krešˇnáková, V ., Nagy, R., Gazdová, J., & Butka, P. 2025, PASP, 137, 034101 Cid Fernandes, R., Stasi ´nska, G., Mateus, A., & Vale Asari, N. 2011, Monthly Notices of the Royal Astronomical Society, 413, 1687 Ciuc˘a, I., Kawata, D., Ting, Y .-S., et al. 20...

    work page 2020

  13. [13]

    O., Scaife, A

    Clarke, A. O., Scaife, A. M. M., Greenhalgh, R., & Griguta, V . 2020, A&A, 639, A84

    work page 2020

  14. [14]

    W., & Dambre, J

    Dieleman, S., Willett, K. W., & Dambre, J. 2015, MNRAS, 450, 1441

    work page 2015

  15. [15]

    2025, A&A, accepted (Euclid Q1 SI), arXiv:2503.15311

    Dobrycheva, D. V ., Hetmantsev, O. O., Vavilova, I. B., et al. 2025b, VizieR On- line Data Catalog: Catalogue of inspected PRGs - CIPRG (Dobrycheva+, 2025), VizieR On-line Data Catalog: J/A+A/702/A258. Originally published in: 2025A&A...702A.258D Domínguez Sánchez, H., Margalef, B., Bernardi, M., & Huertas-Company, M. 2022, MNRAS, 509, 4024 Euclid Collabo...

    work page arXiv 2025

  16. [16]

    N., Primini, F

    Evans, I. N., Primini, F. A., Glotfelty, C. S., et al. 2019, VizieR Online Data Cata- log: The Chandra Source Catalog (CSC), Release 2.0 (Evans+, 2019), VizieR On-line Data Catalog: IX/57. Originally published in: 2010ApJS..189...37E

    work page 2019

  17. [17]

    A., Willett, K

    Galloway, M. A., Willett, K. W., Fortson, L. F., & et al. 2015, Monthly Notices of the Royal Astronomical Society, 448, 3442

    work page 2015

  18. [18]

    M., Lim, D

    Goh, K. M., Lim, D. H. Y ., Sham, Z. D., & Prakash, K. B. 2025, Research in Astronomy and Astrophysics, 25, 065018

    work page 2025

  19. [19]

    2025, A&A, 699, A232

    Heesters, N., Chemaly, D., Müller, O., et al. 2025, A&A, 699, A232

    work page 2025

  20. [20]

    W., Kelvin, L., Baldry, I., et al

    Holwerda, B. W., Kelvin, L., Baldry, I., et al. 2019, AJ, 158, 103

    work page 2019

  21. [21]

    & Lanusse, F

    Huertas-Company, M. & Lanusse, F. 2023, PASA, 40, e001

    work page 2023

  22. [22]

    M., Tremonti, C., et al

    Kauffmann, G., Heckman, T. M., Tremonti, C., et al. 2003, Monthly Notices of the Royal Astronomical Society, 346, 1055

    work page 2003

  23. [23]

    J., Dopita, M

    Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., & Trevena, J. 2001, The Astrophysical Journal, 556, 121

    work page 2001

  24. [24]

    J., Groves, B., Kauffmann, G., & Heckman, T

    Kewley, L. J., Groves, B., Kauffmann, G., & Heckman, T. 2006, Monthly Notices of the Royal Astronomical Society, 372, 961

    work page 2006

  25. [25]

    B., Dobrycheva, D

    Khramtsov, V ., Vavilova, I. B., Dobrycheva, D. V ., et al. 2022, Space Science and Technology, 28, 27

    work page 2022

  26. [26]

    Kim, M., Choi, Y .-Y ., & Kim, S. S. 2020, Monthly Notices of the Royal Astro- nomical Society, 494, 5839

    work page 2020

  27. [27]

    2024, arXiv e-prints, arXiv:2411.10373

    Li, F., Rahman, M., Murray, N., et al. 2024, arXiv e-prints, arXiv:2411.10373

    work page arXiv 2024

  28. [28]

    J., Schawinski, K., Slosar, A., et al

    Lintott, C. J., Schawinski, K., Slosar, A., et al. 2008, MNRAS, 389, 1179

    work page 2008

  29. [29]

    W., Higley, A

    Lyke, B. W., Higley, A. N., McLane, J. N., et al. 2020, ApJS, 250, 8

    work page 2020

  30. [30]

    2025, Research in Astronomy and Astrophysics, 25, 085010

    Mao, Y ., Tu, L., Xu, Z., Jiang, Y ., & Zheng, M. 2025, Research in Astronomy and Astrophysics, 25, 085010

    work page 2025

  31. [31]

    C., & Geach, J

    Martin, G., Kaviraj, S., Hocking, A., Read, S. C., & Geach, J. E. 2020, MNRAS, 491, 1408

    work page 2020

  32. [32]

    2025, arXiv e-prints, arXiv:2502.09610

    Masters, K. 2025, arXiv e-prints, arXiv:2502.09610

    work page arXiv 2025

  33. [33]

    2024, A&A, 682, A34

    Merloni, A., Lamer, G., Liu, T., et al. 2024, A&A, 682, A34

    work page 2024

  34. [34]

    Merrow, A., Grand, R. J. J., Fragkoudi, F., & Martig, M. 2024, MNRAS, 531, 1520

    work page 2024

  35. [35]

    V ., Smirnova, K

    Moiseev, A. V ., Smirnova, K. I., Smirnova, A. A., & Reshetnikov, V . P. 2011, MNRAS, 418, 244

    work page 2011

  36. [36]

    2024, MNRAS, 533, 292

    Mukundan, K., Nair, P., Bailin, J., & Li, W. 2024, MNRAS, 533, 292

    work page 2024

  37. [37]

    Nair, P. B. & Abraham, R. G. 2010, ApJS, 186, 427

    work page 2010

  38. [38]

    B., et al

    Oh, K., Koss, M., Markwardt, C. B., et al. 2018, ApJS, 235, 4

    work page 2018

  39. [39]

    S., et al

    Ren, J., Li, N., Liu, F. S., et al. 2023, ApJ, 958, 96 Sánchez-Menguiano, L., Martínez-Valpuesta, I., & Sánchez, S. F. 2015, Monthly Notices of the Royal Astronomical Society, 450

    work page 2023

  40. [40]

    A., Martins, L

    Silva-Lima, L. A., Martins, L. P., Coelho, P. R. T., & Gadotti, D. A. 2022, As- tronomy & Astrophysics, 661, A105

    work page 2022

  41. [41]

    2024, MNRAS, 531, 4336

    Somawanshi, D., Bhattacharya, S., Kataria, M., & Kobayashi, C. 2024, MNRAS, 531, 4336

    work page 2024

  42. [42]

    Sormani, M. C. & Barnes, A. T. 2019, MNRAS, 484, 1213

    work page 2019

  43. [43]

    C., Treß, R

    Sormani, M. C., Treß, R. G., Glover, S. C. O., et al. 2019, MNRAS, 488, 4663

    work page 2019

  44. [44]

    2024, ApJ, 971, L6

    Su, Y ., Zhang, S., Sun, Y ., et al. 2024, ApJ, 971, L6

    work page 2024

  45. [45]

    2025, A&A, 695, A106

    Thuruthipilly, H., Junais, Koda, J., et al. 2025, A&A, 695, A106

    work page 2025

  46. [46]

    & Shamir, L

    Timmis, I. & Shamir, L. 2017, ApJS, 231, 2

    work page 2017

  47. [47]

    2025, MNRAS, 540, 3493

    Tsukui, T., Wisnioski, E., Bland-Hawthorn, J., & Freeman, K. 2025, MNRAS, 540, 3493

    work page 2025

  48. [48]

    B., Dobrycheva, D

    Vavilova, I. B., Dobrycheva, D. V ., Vasylenko, M. Y ., et al. 2021b, VizieR On- line Data Catalog: SDSS galaxies morphological classification (Vavilova+, 2021), VizieR On-line Data Catalog: J/A+A/648/A122. Originally published in: 2021A&A...648A.122V

    work page 2021

  49. [49]

    B., Khramtsov, V ., Dobrycheva, D

    Vavilova, I. B., Khramtsov, V ., Dobrycheva, D. V ., et al. 2023, VizieR Online Data Catalog (other), 0710, J/other/KNIT/28

    work page 2023

  50. [50]

    F., Salo, H., et al

    Venhola, A., Peletier, R. F., Salo, H., et al. 2022, A&A, 662, A43

    work page 2022

  51. [51]

    2022, MNRAS, 509, 3966

    Walmsley, M., Lintott, C., Géron, T., et al. 2022, MNRAS, 509, 3966

    work page 2022

  52. [52]

    2022, PASP, 134, 114508

    Wei, S., Li, Y ., Lu, W., et al. 2022, PASP, 134, 114508

    work page 2022

  53. [53]

    W., Lintott, C

    Willett, K. W., Lintott, C. J., Bamford, S. P., et al. 2013, MNRAS, 435, 2835

    work page 2013

  54. [54]

    S., et al

    Xu, Q., Shen, S., de Souza, R. S., et al. 2023, MNRAS, 526, 6391

    work page 2023

  55. [55]

    & Dalcanton, J

    Yoachim, P. & Dalcanton, J. J. 2005, ApJ, 624, 701

    work page 2005

  56. [56]

    Zaw, I., Chen, Y .-P., & Farrar, G. R. 2019, ApJ, 872, 134

    work page 2019

  57. [57]

    Z., Hafezianzadeh, F., Zhang, Y ., et al

    Zeraatgari, F. Z., Hafezianzadeh, F., Zhang, Y ., et al. 2024, MNRAS, 527, 4677 Article number, page 10 of 10

    work page 2024