Pith. sign in

REVIEW 4 major objections 6 minor 34 references

SPACE-SUIT: An Artificial Intelligence Based Chromospheric Feature Extractor and Classifier for SUIT

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read AI detector finds four chromospheric features in SUIT images

desk verdict Useful engineering for SUIT, but the headline mock-data numbers need a documented source-level split before they can be trusted, and the real-data claim is qualitative. read the letter →

arxiv 2412.08589 v2 pith:PYB4N74R submitted 2024-12-11 astro-ph.SR astro-ph.IMcs.CVcs.LG

Pranava Seth , Vishal Upendran , Megha Anand , Janmejoy Sarkar , Soumya Roy , Priyadarshan Chaki , Pratyay Chowdhury , Borishan Ghosh
show 1 more author
Durgesh Tripathi
This is my paper · ORCID
classification astro-ph.SRastro-ph.IMcs.CVcs.LG
keywords chromospheresunspotdetectionplageregionssolarfilamentsoff-limbstructuresobjectYOLOTamurafeatures
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

SPACE-SUIT is an automatic detector that classifies four chromospheric structures—plage regions, sunspots, filaments, and off-limb structures—in images from SUIT, the ultraviolet telescope on India's Aditya-L1 observatory. Because SUIT had not accumulated enough labeled images, the authors train the detector on 'mock-SUIT' images synthesized from ten years of IRIS Mg II k full-disk mosaics, folded through SUIT's effective area and point-spread function. On the mock validation data the best model, a YOLOv8 oriented-bounding-box network, reaches a precision of about 0.788, a recall of about 0.863, and a mean average precision of about 0.874; on real SUIT Level-1 images it detects most structures of interest after a rescaling adjustment, but this evaluation is qualitative. The paper also introduces a label-free 'self-validation' scheme that compares statistical and Tamura texture features of ground-truth and predicted regions, arguing this can stand in for human labels when none exist.

What carries the argument

The central mechanism is a supervised object-detection pipeline built on YOLOv8 with oriented bounding boxes (OBB), which predicts rotated rectangles for features of arbitrary orientation. The training data is mock-SUIT: roughly ten years of IRIS Mg II k full-disk mosaics are wavelength-summed over SUIT's NB3 effective area, cleaned to remove cosmic rays and mosaic stitching artifacts, convolved with SUIT's NB3 point-spread function, and rebinned to SUIT plate scale, producing images that stand in for real SUIT observations. The validation mechanism is a set of ten scalar descriptors per bounding box—Shannon entropy, standard deviation, skewness, kurtosis, and five gray-level co-occurrence matrix statistics (contrast, homogeneity, dissimilarity, energy, correlation)—whose distributions are compared across ground truth, predictions, and real SUIT detections.

What would settle it

Manually label a few hundred real SUIT NB3 full-disc images, run SPACE on them, and compute the per-class precision and recall; if the numbers fall well below the mock-SUIT validation values (precision ≈0.788, recall ≈0.863), the mock-training transfer assumption is false.

Watch

Extended reading notes

Core claim

The paper's central claim is that SPACE—a YOLOv8-obb network trained on mock-SUIT images—can detect and classify plage regions, sunspots, filaments, and off-limb structures in SUIT's Mg II k images with a precision of ≈0.788, a recall of ≈0.863, and a mean average precision of ≈0.874 on the mock validation FITS dataset, and can identify many of these structures in real SUIT Level-1 data. A second claim is that distributions of statistical and Tamura features computed inside detected bounding boxes agree between ground truth and predictions on mock data and differ across the four feature classes, providing a way to self-validate detections where no labels exist.

Load-bearing premise

The load-bearing premise is that mock-SUIT images—IRIS Mg II k mosaics convolved with SUIT's effective area and PSF—are close enough to real SUIT NB3 observations that a detector trained on them works without retraining; the paper's own rescaling and fine-tuning caveats show this premise holds only approximately.

Editorial extensions

If this is right

  • SPACE itself is a candidate Level 2+ data product for SUIT: it can produce per-image catalogs of plages, sunspots, filaments, and off-limb structures automatically and in near real time.
  • Because the model is trained on synthetic data, the same recipe can be rerun as SUIT accumulates real images, with the labeled real data replacing or augmenting the mock set.
  • The reported per-class scores imply that plages and filaments are detected with high recall (≈0.98 and ≈0.96 on FITS validation) while off-limb structures are the weak class (recall ≈0.54), so the model's practical use should start with the well-detected classes.
  • The self-validation descriptors separate the four classes (for example, sunspots have low entropy and high dissimilarity, filaments high entropy), so they can flag predictions whose texture does not match the training distribution, as happens for off-limb regions on real SUIT data.
  • Deploying SPACE on real data currently requires a manual rescaling of the image intensities by a factor of about 8, and the authors state that fine-tuning on real SUIT data is needed for effective identification.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension would be to train on a small set of hand-labeled real SUIT images and measure how much the 0.788/0.863 validation numbers change; if the gap closes, the mock-to-real transfer can be solved by fine-tuning rather than by a deeper domain model.
  • The self-validation distributions could double as an instrument-health monitor: since SUIT's throughput drifts with contaminant deposition and baking cycles, tracking entropy or contrast statistics over time may flag when a detector retrain is needed.
  • If the mock-image generator is updated to match SUIT's actual PSF, flat-field, and stray-light residuals, the same training recipe could be transferred to other Mg II k imagers (or other SUIT bandpasses) without re-architecting the detector.
  • The off-limb recall deficit (0.542) suggests that elongated, limb-crossing structures are underrepresented or too heterogeneous in the OBB labels; augmenting with more off-limb examples or a segmentation head might be needed for complete catalogs.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper presents SPACE-SUIT, a YOLOv8-obb-based object detector for identifying plages, sunspots, filaments, and off-limb structures in SUIT Mg II k (NB3) images. Because few real SUIT images are available, the authors construct mock-SUIT images from IRIS Mg II k full-disk mosaics by convolving with the SUIT NB3 effective area and PSF, manually label bounding boxes, and train three YOLO variants (v8, v8-obb, v9). The v8-obb model is selected as best, with reported validation metrics on mock FITS data of precision 0.788, recall 0.863, and MAP@0.5 0.874 (Table 3). The model is then applied to real SUIT Level-1 images using an ad hoc rescaling factor of 8, with qualitative comparisons to AIA 1700 Å sunspots. The paper also introduces a 'self-validation' scheme based on statistical moments and Tamura texture features computed inside detected and ground-truth bounding boxes, arguing that consistency between these distributions supports the detections on real data. The paper concludes that entropy best discriminates the four feature classes and that the method demonstrates 'zero-shot' capability on SUIT data, while acknowledging that fine-tuning on real SUIT data would be needed.

Significance. If the mock-data metrics are trustworthy, this is a practically useful contribution: it provides a ready-to-use detector for a new solar mission (SUIT) and demonstrates a transfer pipeline from IRIS mosaics to SUIT-like images. The comparative evaluation of YOLO v8/v8-obb/v9 and the public description of the mock-image generation are useful for future instrument-specific detectors. The self-validation idea using statistical and Tamura features is interesting as a qualitative diagnostic when no ground truth exists, though it is not a substitute for an independent test set. The main weakness is that the central quantitative claim (the Table 3 numbers) depends on a train/validation split that is not described, and the real-SUIT validation is only qualitative and involves a manually tuned rescaling. These issues are fixable, so the work is potentially acceptable after substantial revision.

major comments (4)
  1. [Section 2.2, Table 1] The manuscript never describes how the 217 pre-augmentation images were split into training and validation sets. Because the training set is augmented to 3056 images and the validation set includes 123 FITS images, it is unknown whether any validation image is an augmented or PSF-convolved variant of a training mosaic. The headline metrics in Table 3 (precision 0.788, recall 0.863, MAP 0.874) are only meaningful if the validation images come from IRIS mosaics that are disjoint from all training mosaics. The authors must specify the source-level split (e.g., temporal, by observation date, or by random mosaic assignment) and ideally report performance on a held-out set of IRIS observations that were never used in training.
  2. [Section 5.1.2, Figure 6] The real-SUIT evaluation is not quantitative and the preprocessing is tuned on the same images: the rescaling factor of 8 is chosen by 'trial and error' on the SUIT images themselves (Section 5.1.2). The later statement in Section 6 that the distributions give 'confidence in the performance of our model on unseen data' and support 'zero-shot learning' is therefore not supported. The authors should either provide a quantitative comparison of sunspot detections against AIA 1700 Å (e.g., overlap, centroid error, or detection rate) using a fixed, pre-registered preprocessing rule, or explicitly restrict the claims to a qualitative demonstration and present the rescaling factor as a limitation requiring fine-tuning.
  3. [Sections 4, 5.2, and 6] The self-validation scheme computes distributions of statistical and Tamura features on the same training and validation splits used to train the model. This procedure can only show that detected boxes resemble typical training boxes; it cannot detect missed detections, localization errors, or biases that preserve texture statistics. The paper partially acknowledges this in Section 6 ('It does not... provide a measure of the fraction of all regions captured correctly'), but the abstract and the phrase 'validated with the observed SUIT images' overstate the evidence. The authors should soften these claims and make clear that the self-validation is a consistency check, not an independent validation.
  4. [Section 2.2, Eq. (1)] The gamma augmentation is selected by 'visual inspection' with only the 'best looking' values used, and the same subsection states that the full range of gamma is not used. This introduces a manually chosen, data-dependent free parameter. The authors should either justify this selection more rigorously (e.g., by a sensitivity analysis of model performance versus gamma range) or otherwise treat it as a hyperparameter that is reported in the final model configuration.
minor comments (6)
  1. [Section 2.2] The text says the data are saved as 'lossless JPG images,' but JPEG is a lossy format. This is also inconsistent with the later statement in Section 5.1.1 that JPG data are 'lossy' and quantized. Please reword to 'high-quality JPEG' or use a lossless format such as PNG.
  2. [Section 2.2, Table 1] The text says the original images are 'approximately 200' while Table 1 lists 217 pre-augmentation images; please harmonize these numbers.
  3. [Section 5.1.2 and Section 2] The manuscript refers to 'Level 1' data in the abstract and Table 1 but mentions 'level 1.1 SUIT data' in Section 5.1.2; please clarify which data level is used and whether the calibration steps differ.
  4. [Figures 7 and 8] The captions say the rows correspond to the listed measures, but the text in Section 5.2 refers to 'first row,' 'second row,' etc.; consider numbering the subplots or explicitly referring to panels to improve readability.
  5. [Table 4] The row 'Test (SUIT Data)' shows 'NA' for filaments; please explain why no filaments were detected in the real SUIT images, as this is relevant to the claim of detecting all four feature classes.
  6. [References] The reference for Schuh and Angryk (2014) contains '???' in the venue field; please provide the full bibliographic information.

Circularity Check

2 steps flagged · score 4.0 of 10

Mock-FITS metrics are externally labeled and largely independent, but the real-SUIT 'validation' mixes a test-set-tuned rescaling factor with a self-referential statistical check, so the real-data confirmation is partly circular.

  1. fitted input called prediction [Section 5.1.2 (Model Deployed on the Real SUIT Data), Figure 6]
    "We perform a trial and error of rescaling, and find that a rescaling of each image by a factor of 8 helps in better identification of the features in the SUIT images. These estimated boxes are shown in the second row of Figure 6."

    The paper frames the real-data exercise as a 'zero-shot prediction on the SUIT observations' (Section 1), but the rescaling factor of 8 was obtained by trial and error on those same observations. The displayed detections therefore depend on a parameter fitted to the evaluation data; the 'prediction' is not made before seeing the test set, so the qualitative success on real SUIT is partly constructed rather than independently confirmed. The authors disclose the manual rescaling, but the framing as zero-shot validation overstates what this step demonstrates.

  2. other [Abstract; Section 5.2 and Section 6 (self-validation with statistical and Tamura measures)]
    "Since our dataset is manually labeled, we perform 'self-validation' by applying statistical measures and Tamura features on the ground truth and predicted bounding boxes."

    The statistical/Tamura 'self-validation' compares the model's own predicted boxes with distributions derived from the manual labels that supervised the training. Because the model was optimized to reproduce those labels, agreement between ground-truth and predicted distributions on mock data is a consistency check rather than an independent test. Extending the same training-derived distributions to SUIT detections cannot certify correctness without labeled ground truth, as the authors acknowledge when they say fine-tuning on real SUIT data is still needed. This step is explicitly labeled self-validation, so it is a disclosed limitation rather than a hidden derivation of the headline mock-FITS metrics.

full rationale

The central quantitative claim, precision approx 0.788, recall approx 0.863, MAP approx 0.874 on the validation mock FITS set (Table 3), is evaluated against manual bounding-box labels and is therefore not circular by construction. The mock-SUIT construction from IRIS mosaics and the training/evaluation split are presented as a standard supervised pipeline. The main circular components are confined to the real-SUIT demonstration: the rescaling factor is tuned on the test observations before the detections are shown as successful, and the 'self-validation' scheme uses distributions derived from the same labels that trained the model. The paper is transparent about both aspects ('self-validation', 'trial and error of rescaling', 'fine-tuning on real SUIT data'), which keeps the issue proportionate. A separate correctness risk, not counted as demonstrated circularity, is that Section 2.2/Table 1 never states that the 123 FITS validation samples come from IRIS mosaics disjoint from the 217 pre-augmentation training sources; if validation images are augmented or PSF-convolved variants of training sources, the headline metrics could measure memorization. That possibility is not established by a quoted reduction, so it is raised only as a risk. Overall, the paper's core mock-data evaluation has independent content; the circularity score is moderate because the real-data validation is partly self-referential and test-tuned.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claims rest on the transferability of mock-SUIT training data, the correctness of manual labels, and the discriminative power of the texture statistics. No new physical entities are introduced. The main free parameter is the post-hoc rescaling factor applied to real SUIT images.

free parameters (2)
  • SUIT image rescaling factor = 8
    Applied to real SUIT Level-1 images before inference; chosen by trial and error in Section 5.1.2 because min-max scaling caused the model to miss sunspots. This is post-hoc tuning on the test data.
  • Gamma augmentation subset = subset of gamma values 0.2, 0.4, 0.6, 0.8
    Section 2.2: only augmented images that 'do not look starkly different' from source images were retained, based on visual inspection. This manual selection is a form of hand-picking training data.
assumptions (3)
  • domain assumption Mock-SUIT images are representative of real SUIT NB3 observations
    Section 2.1 generates mock images from IRIS mosaics using SUIT effective area and PSF; the paper's own Section 6 notes large differences in intensity statistics between mock and real SUIT data.
  • domain assumption Manual bounding-box labels are reliable ground truth
    Section 2.2: one round of hand-labeling via Roboflow with no described quality control, inter-labeler agreement, or uncertainty.
  • domain assumption Statistical and Tamura features distinguish the four feature classes
    Section 4 defines these features as a self-validation tool; Section 5.2 demonstrates separability only by visual histogram comparison, not by a quantitative classifier or statistical test.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SPACE-SUIT: An Artificial Intelligence Based Chromospheric Feature Extractor and Classifier for SUIT." pith.science (2026). https://pith.science/paper/PYB4N74R

@misc{pith2026241208589,
  author       = {Pith},
  title        = {Pith review of: SPACE-SUIT: An Artificial Intelligence Based Chromospheric Feature Extractor and Classifier for SUIT},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PYB4N74R}},
  note         = {Machine review of arXiv:2412.08589}
}
read the original abstract

The Solar Ultraviolet Imaging Telescope(SUIT) onboard Aditya-L1 is an imager that observes the solar photosphere and chromosphere through observations in the wavelength range of 200-400 nm. A comprehensive understanding of the plasma and thermodynamic properties of chromospheric and photospheric morphological structures requires a large sample statistical study, necessitating the development of automatic feature detection methods. To this end, we develop the feature detection algorithm SPACE-SUIT: Solar Phenomena Analysis and Classification using Enhanced vision techniques for SUIT, to detect and classify the solar chromospheric features to be observed from SUIT's Mg II k filter. Specifically, we target plage regions, sunspots, filaments, and off-limb structures. SPACE uses YOLO, a neural network-based model to identify regions of interest. We train and validate SPACE using mock-SUIT images developed from Interface Region Imaging Spectrometer(IRIS) full-disk mosaic images in Mg II k line, while we also perform detection on Level-1 SUIT data. SPACE achieves an approximate precision of 0.788, recall 0.863 and MAP of 0.874 on the validation mock SUIT FITS dataset. Given the manual labeling of our dataset, we perform "self-validation" by applying statistical measures and Tamura features on the ground truth and predicted bounding boxes. We find the distributions of entropy, contrast, dissimilarity, and energy to show differences in the features. These differences are qualitatively captured by the detected regions predicted by SPACE and validated with the observed SUIT images, even in the absence of labeled ground truth. This work not only develops a chromospheric feature extractor but also demonstrates the effectiveness of statistical metrics and Tamura features for distinguishing chromospheric features, offering independent validation for future detection schemes.

Figures

Figures reproduced from arXiv: 2412.08589 by the authors.

Figure 1
Figure 1. Effective Area of NB3 filter of SUIT normalized by its peak, coinciding with the wavelength range IRIS Mg II k mosaics. The green and red vertical lines specify lower and higher wavelength limits recorded by IRIS, respectively. SPACE is a supervised learning algorithm, that needs pairs of input images, bounding boxes. These input images, on deployment of the workflow, would be the SUIT images. As of this manuscript,… view at source ↗
Figure 2
Figure 2. We present the preprocessing of IRIS mosaics to mock-SUIT data for two examples, presented in individual columns. From Top to Bottom: Each row represents a step in the image processing pipeline. (a) IRIS mosaic pixel values at the central wavelength of 2796 ˚A. (b) The same IRIS mosaics folded with the normalized NB3 effective area (bottom row). (c) Image with cosmic ray artifacts removed using a 5×5 median filter. … view at source ↗
Figure 3
Figure 3. Continued from [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Example representative augmentation for an example mock SUIT image. Starting from the top-left and moving clockwise, the transformations include: Gray Scale conversion, HSV conversion, γ = 0.4, vertical and horizontal flip, and γ = 0.6 [PITH_FULL_IMAGE:figures/full_fi…
Figure 5
Figure 5. Figure 5: Predictions on SUIT MOCK FITS. F: Filaments, OL: Off Limb features, S: Sunspots P: Plages. The numbers in each box present the confidence score calculated for each feature between 0 and 1. The images shown were observed on 13 October 2013 (right column) and 9 September…
Figure 6
Figure 6. Figure 6: Inference on SUIT Images (top two panels) and reference AIA 1700 images (bottom panel). The features are given as: F: Filaments, OL: Off Limb features, S: Sunspots P: Plages; Confidence Score calculated for each event, whose value lies between 0 and 1. The top row corr…
Figure 7
Figure 7. Figure 7: Variation of the statistical measures for the features of interest, in terms of entropy, σ, skew and kurtosis along each row. The columns correspond to the train ground truth, train prediction, validation ground truth, validation prediction, and SUIT prediction respect…
Figure 8
Figure 8. Figure 8: Variation of the Tamura measures for the features of interest, including con￾trast, dissimilarity, homogeneity, energy, and correlation. The columns correspond to the train ground truth, train prediction, validation ground truth, validation prediction, and SUIT predict…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 25 canonical work pages

  1. [1]

    In:2017 IEEE International Conference on Big Data (Big Data),

    Ahmadzadeh, A., Kempton, D.J., Schuh, M.A., Angryk, R.A.: 2017, Improving the function- ality of tamura directionality on solar images. In:2017 IEEE International Conference on Big Data (Big Data),

    work page 2017

  2. [2]

    In:Pro- ceedings of the 18th ACM International Conference on Multimedia,MM ’10, Association for Computing Machinery, New York, NY, USA, 1485–1488

    Marcel, S., Rodriguez, Y.: 2010, Torchvision the machine-vision package of torch. In:Pro- ceedings of the 18th ACM International Conference on Multimedia,MM ’10, Association for Computing Machinery, New York, NY, USA, 1485–1488. ISBN 9781605589336. DOI. https://doi.org/10.1145/1873951.1874254. Mumford, S., Freij, N., Christe, S., Ireland, J., Mayer, F., H...

    arXiv 2010

  3. [3]

    IEEE. Barata, T., Carvalho, S., Dorotoviˇ c, I., Pinheiro, F.J.G., Garcia, A., Fernandes, J., Louren¸ co, A.M.: 2018, Software tool for automatic detection of solar plages in the Coimbra Observatory spectroheliograms.Astronomy and Computing24,

    work page 2018

  4. [10]

    DOI. ADS. Bradski, G.: 2000, The OpenCV Library.Dr. Dobb’s Journal of Software Tools. Chen, R.: 2016, A Brief Introduction on Shannon’s Information Theory. DOI. Clark, A.: 2015,Pillow (PIL Fork) Documentation, readthedocs. https://buildmedia. readthedocs.org/media/pdf/pillow/latest/pillow.pdf. Colak, T., Qahwaji, R.: 2008, Automated McIntosh-Based Classif...

    work page 2000

  5. [13]

    Hunter, J.D.: 2007, Matplotlib: A 2D graphics environment.Computing in science & engineering9,

    DOI. Hunter, J.D.: 2007, Matplotlib: A 2D graphics environment.Computing in science & engineering9,

    work page 2007

  6. [17]

    https://doi.org/10.1007/s11207-011-9776-8

    DOI. https://doi.org/10.1007/s11207-011-9776-8. Liu, D., Song, W., Lin, G., Wang, H.: 2021, Solar Filament Segmentation Based on Improved U-Nets.Solar Physics296,

    doi:10.1007/s11207-011-9776-8 2021

  7. [22]

    DOI. ADS. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Kopf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., Chintala, S.: 2019, PyTorch: An Imperative Style, High-Performance Deep Learning Library. In:Advances in ...

    work page 2019

  8. [24]

    DOI. ADS. Price-Whelan, A.M., Sip˝ ocz, B., G¨ unther, H., Lim, P., Crawford, S., Conseil, S., Shupe, D., Craig, M., Dencheva, N., Ginsburg, A., et al.: 2018, The Astropy Project: Building an open-science project and status of the v2. 0 core package.The Astronomical Journal156,

    work page 2018

Show all 34 references
  1. [30]

    DOI. van der Walt, S., Sch¨ onberger, J.L., Nunez-Iglesias, J., Boulogne, F., Warner, J.D., Yager, N., Gouillart, E., Yu, T., the scikit-image contributors: 2014, scikit-image: image processing in Python.PeerJ2, e453. DOI. http://dx.doi.org/10.7717/peerj.453. Veeramani, M., Su...

    2014 doi

  2. [33]

    https://aip.scitation.org/doi/abs/10.1109/MCSE.2011.37

    DOI. https://aip.scitation.org/doi/abs/10.1109/MCSE.2011.37. Wang, C.-Y., Yeh, I.-H., Liao, H.-Y.M.: 2024,YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information. Waskom, M., Botvinnik, O., O’Kane, D., Hobson, P., Lukauskas, S., Gemperline, D.C., Augspu...

    2011 doi

  3. [36]

    Kempton, D.J., Schuh, M.A., Angryk, R.A.: 2018, Tracking Solar Phenomena from the SDO

    DOI. Kempton, D.J., Schuh, M.A., Angryk, R.A.: 2018, Tracking Solar Phenomena from the SDO. The Astrophysical Journal869,

    2018

  4. [41]

    DOI. ADS. Bourgeois, S., Barata, T., Erd´ elyi, R., Gafeira, R., Oliveira, O.: 2024, Sunspots Identification Through Mathematical Morphology.Solar Physics299,

    2024

  5. [54]

    DOI. ADS. Kido, H.: 2023, Image Inpainting Methods. Digital Image Reconstruction and Restoration. Lemen, J.R., Title, A.M., Akin, D.J., Boerner, P.F., Chou, C., Drake, J.F., Duncan, D.W., Edwards, C.G., Friedlaender, F.M., Heyman, G.F., Hurlburt, N.E., Katz, N.L., Kushner, G.D...

    2023

  6. [60]

    https://link.springer.com/10.1007/s11207-025-02468-2

    DOI. https://link.springer.com/10.1007/s11207-025-02468-2. Schuh, M., Angryk, R.: 2014, Massive Labeled Solar Image Data Benchmarks for Automated Feature Recognition. In:???DOI. Schuh, M.A., Angryk, R.A., Pillai, K.G., Banda, J.M., Martens, P.C.: 2013, A large-scale solar imag...

    2014 doi

  7. [62]

    Padinhatteeri, S., Higgins, P.A., Bloomfield, D.S., Gallagher, P.T.: 2016, Automatic Detection of Magneticδin Sunspot Groups.Solar Physics291,

    DOI. Padinhatteeri, S., Higgins, P.A., Bloomfield, D.S., Gallagher, P.T.: 2016, Automatic Detection of Magneticδin Sunspot Groups.Solar Physics291,

    2016

  8. [70]

    DOI. ADS. Benkhalil, A., Zharkova, V.V., Zharkov, S., Ipson, S.: 2006, Active Region Detection and Verification With the Solar Feature Catalogue.Solar Physics235,

    2006

  9. [87]

    DOI. ADS. Bertalmio, M., Bertozzi, A.L., Sapiro, G.: 2001, Navier-stokes, fluid dynamics, and image and video inpainting. In:Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 20011, I. DOI. SOLA: manuscript.tex; 22 Septem...

    2001

  10. [90]

    https://github.com/ultralytics/ ultralytics

    Jocher, G., Chaurasia, A., Qiu, J.: 2023,Ultralytics YOLOv8. https://github.com/ultralytics/ ultralytics. Kempton, D., Pillai, K.G., Angryk, R.: 2014, Iterative refinement of multiple targets tracking of solar events. In:2014 IEEE International Conference on Big Data (Big Data),

    2023

  11. [101]

    DOI. ADS. Zhao, C., Lin, G., Deng, Y., Yang, X.: 2016, Automatic Recognition of Sunspots in HSOS Full- Disk Solar Images.Publications of the Astronomical Society of Australia33, e018. DOI. ADS. SOLA: manuscript.tex; 22 September 2025; 17:28; p. 26

    2016

  12. [104]

    DOI. ADS. Hall-Beyer, M.: 2017, GLCM Texture: A Tutorial v. 3.0 March

    2017

  13. [119]

    DOI. ADS. Ren, S., He, K., Girshick, R., Sun, J.: 2016,Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. https://arxiv.org/abs/1506.01497. Sarkar, J., Nived, V.N., Roy, S., Deogaonkar, R., Padinhatteeri, S., Bayanna, A.R., Ke- sharwani, R., Ramapr...

    2016 arXiv

  14. [123]

    Qu, M., Shih, F.Y., Jing, J., Wang, H.: 2005, Automatic Solar Filament Detection Using Image Processing Techniques.Solar Physics228,

    2005

  15. [176]

    DOI. ADS. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., Berg, A.C.: 2016,SSD: Sin- gle Shot MultiBox Detector, Springer International Publishing, 21–37. ISBN 9783319464480. DOI. http://dx.doi.org/10.1007/978-3-319-46448-0

    2016 doi

  16. [193]

    DOI. ADS. Walt, S.v.d., Colbert, S.C., Varoquaux, G.: 2011, The NumPy Array: A Structure for Efficient Numerical Computation.Computing in Science & Engineering13,

    2011

  17. [261]

    DOI. ADS. ˇSimberov´ a, S., Karlick´ y, M., Suk, T.: 2014, Statistical Moments of Active-Region Images During Solar Flares.Solar Physics289,

    2014

  18. [277]

    DOI. ADS. De Pontieu, B., Lemen, J., Kushner, G., Akin, D., Allard, B., Berger, T., Boerner, P., Cheung, M., Chou, C., Drake, J., et al.: 2014, The interface region imaging spectrograph (IRIS). Solar Physics289,

    2014

  19. [616]

    http://dx.doi.org/10.18520/cs/v113/i04/616-619

    DOI. http://dx.doi.org/10.18520/cs/v113/i04/616-619. SOLA: manuscript.tex; 22 September 2025; 17:28; p. 25 Seth et al. Tripathi, D., Chakrabarty, D., Nandi, A., Prasad, B., Ramaprakash, A., Shaji, N., Sankarasub- ramanian, K., Thampi, R., Yadav, V.: 2023, The Aditya-L1 mission...

    2025 doi

  20. [1413]

    DOI. ADS. Guo, X., Yang, Y., Feng, S., Bai, X., Liang, B., Dai, W.: 2022, Solar-Filament Detection and Classification Based on Deep Learning.Solar Physics297,

    2022

  21. [1832]

    DOI. ADS. Otsu, N.: 1979, A Threshold Selection Method from Gray-Level Histograms.IEEE Transac- tions on Systems, Man, and Cybernetics9,

    1979

  22. [2017]

    Huang, T., Yang, G., Tang, G.: 1979, A fast two-dimensional median filtering algorithm.IEEE Transactions on Acoustics, Speech, and Signal Processing27,

    DOI. Huang, T., Yang, G., Tang, G.: 1979, A fast two-dimensional median filtering algorithm.IEEE Transactions on Acoustics, Speech, and Signal Processing27,

    1979

  23. [2518]

    Ahmadzadeh, A., Mahajan, S.S., Kempton, D.J., Angryk, R.A., Ji, S.: 2019, Toward Filament Segmentation Using Deep Neural Networks.arXiv e-prints, arXiv:1912.02743

    DOI. Ahmadzadeh, A., Mahajan, S.S., Kempton, D.J., Angryk, R.A., Ji, S.: 2019, Toward Filament Segmentation Using Deep Neural Networks.arXiv e-prints, arXiv:1912.02743. DOI. ADS. Banda, J.M., Liu, C., Angryk, R.A.: 2013, Region-based querying of solar data using descriptor sig...

    2019 arXiv

  24. [2733]

    DOI. Diercke, A., Jarolim, R., Kuckein, C., Gonz´ alez Manrique, S.J., Ziener, M., Veronig, A.M., Denker, C., P¨ otzi, W., Podladchikova, T., Pevtsov, A.A.: 2024, A universal method for solar filament detection from Hαobservations using semi-supervised deep learning.Astronomy ...

    2024

  25. [4349]

    Tan, M., Pang, R., Le, Q.V.: 2020,EfficientDet: Scalable and Efficient Object Detection

    IEEE. Tan, M., Pang, R., Le, Q.V.: 2020,EfficientDet: Scalable and Efficient Object Detection. https://arxiv.org/abs/1911.09070. Terven, J., C´ ordova-Esparza, D.-M., Romero-Gonz´ alez, J.-A.: 2023, A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 t...

    2020 arXiv

  26. [8024]

    Pesnell, W.D., Thompson, B.J., Chamberlin, P.C.: 2012, The Solar Dynamics Observatory (SDO).Solar Physics275,

    http://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance- deep-learning-library.pdf. Pesnell, W.D., Thompson, B.J., Chamberlin, P.C.: 2012, The Solar Dynamics Observatory (SDO).Solar Physics275,

    2012

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.