DeepDISC-Euclid: Source Classification and Photometric Redshifts in Euclid Deep Field North With a Pixel-Level Deep Learning Approach

Abstract

The first Euclid Quick Data Release (Q1) provides extensive imaging and spectroscopic data for hundreds of millions of photometric objects across several deep fields. Accurate classifications and photometric redshifts (photo-z) for these sources are crucial to maximizing the value of these data. In this work, we perform source classification and photo-z estimation for the Euclid Deep Field North (EDF-N) around the North Ecliptic Pole, using a deep learning framework (DeepDISC) that learns and infers using 9-band images simultaneously. We train three dedicated models for (1) source detection and classification, (2) galaxy photo-z, and (3) quasar photo-z. The Euclid Q1 input source catalog, and classifications and spectroscopic redshifts (spec-z) from the Dark Energy Spectroscopic Instrument Data Release 1 are adopted as our training data. DeepDISC source detection achieves overall completeness of ~93% and purity of ~80% if using the Euclid source catalog as the ground truth. Using a JWST source catalog within EDF-N as the reference, we estimate a true purity of ~ 90% for DeepDISC sources. About 99.2%, 99.0%, and 84.8% of stars, galaxies, and quasars, respectively, are correctly recovered with their spectroscopic classifications. The DeepDISC photo-zs show good agreement with spectroscopic redshifts, for both galaxies and quasars. Comparisons with other Euclid Q1 products demonstrate that DeepDISC provides comparable or improved performance in source detection/deblending, classification and photo-z, especially for quasars. These results demonstrate the potential of pixel-level deep learning approaches for large-scale sky surveys such as Euclid and Roman, which will continue to improve with better training labels. We release the full DeepDISC source catalog (~13 million objects) for EDF-N with classifications and photo-zs, including photo-z probability distributions.

Figures (17)

• Figure 1: Magnitude (total flux) distributions in Euclid bands ($I_{\mathrm{E}}$, $Y_{\mathrm{E}}$, $J_{\mathrm{E}}$, and $H_{\mathrm{E}}$) for spectroscopically confirmed stars, galaxies, and QSOs, matched between the Euclid MER catalog in EDF-N and DESI DR1. Total fluxes in each band are computed from the 2FWHM aperture flux in the Euclid MER catalog, following the Photometry Cookbook Euclid:Data, and converted to magnitudes for display.
• Figure 2: Redshift distributions of galaxies and QSOs matched between the Euclid MER catalog in EDF-N and DESI DR1.
• Figure 3: (a) The DeepDISC three-model pipeline applied to the Euclid Deep Field North. Nine-band Euclid+UNIONS images are processed by Model 1 for simultaneous detection, instance segmentation, and star/galaxy/QSO classification. Models 2 and 3 append galaxy and QSO photometric redshift PDFs respectively to produce the final merged catalog of $\sim$13 million sources. Dashed lines indicate training label inputs. (b) Staged training strategy. Stage 1 trains all network weights jointly using Euclid segmentation maps and DESI DR1 spectroscopic classifications. Stage 2 initializes Models 2 and 3 from the Stage 1 checkpoint, then freezes the backbone and classification weights and trains only the new MDN photo-$z$ ROI heads on the DESI spec-$z$ sample. Freezing is necessary because the $\sim$42,000 DESI-labeled sources available for photo-$z$ supervision are too sparse to jointly retrain the full network without degrading the detection and deblending features learned from the much larger Euclid segmentation map ground truth.
• Figure 4: Detection recall (completeness) and precision (purity) of Model 1 as functions of $I_{\mathrm{E}}$ on the test set. Blue circles denote recall, and red triangles denote precision. Poisson uncertainties are shown.
• Figure 5: Example comparisons between Euclid (left) and Model 1 (right) segmentation maps. The Euclid segmentation maps are shown in the left panels, while the Model 1–predicted segmentation maps are shown in the right panels. The top row displays cutouts in the $I_{\mathrm{E}}$ band, and the bottom row shows cutouts in the $Y_{\mathrm{E}}$ band. Red solid contours indicate objects detected by both Euclid and Model 1 (ED); blue dashed contours indicate objects detected only by the DeepDISC Model 1 (D); and yellow dotted contours indicate objects detected only by Euclid (E). DESI objects are additionally highlighted with red squares, with classifications from both DESI and Model 1 indicated. Overall, most model detections are consistent with Euclid detections. Unmatched detections primarily arise from faint sources, artifacts, blended systems, or sources that are not visible in all bands.