CSST Strong Lensing Preparation: a Framework for Detecting Strong Lenses in the Multi-color Imaging Survey by the China Survey Space Telescope (CSST)

TL;DR

This work presents an end-to-end framework for detecting strong gravitational lensing systems in the CSST era by combining full-image processing with a hierarchical Swin Transformer and a sliding-window approach. It couples a data-generation pipeline based on CosmoDC2 with a CSST imaging simulator, image-preprocessing steps (cropping, PSF-deconvolution, asinh grayscale), and an accurate detection network that yields high precision and recall on simulated data ($\text{Precision}\approx$ $0.98$, $\text{Recall}\approx$ $0.90$) and demonstrates transfer to real data from DESI and Euclid. Key contributions include the PSF-Net deconvolution framework with a dual-network loss, the asinh-based grayscale transformation to enhance faint features, and a robust Swin Transformer-based detector validated across simulations and multiple surveys. The results imply strong potential for CSST and cross-survey lensing science, while also highlighting limitations from simplified morphologies and guiding future improvements via more realistic training data and joint multi-survey training.

Abstract

Strong gravitational lensing is a powerful tool for investigating dark matter and dark energy properties. With the advent of large-scale sky surveys, we can discover strong lensing systems on an unprecedented scale, which requires efficient tools to extract them from billions of astronomical objects. The existing mainstream lens-finding tools are based on machine learning algorithms and applied to cut-out-centered galaxies. However, according to the design and survey strategy of optical surveys by CSST, preparing cutouts with multiple bands requires considerable efforts. To overcome these challenges, we have developed a framework based on a hierarchical visual Transformer with a sliding window technique to search for strong lensing systems within entire images. Moreover, given that multi-color images of strong lensing systems can provide insights into their physical characteristics, our framework is specifically crafted to identify strong lensing systems in images with any number of channels. As evaluated using CSST mock data based on an Semi-Analytic Model named CosmoDC2, our framework achieves precision and recall rates of 0.98 and 0.90, respectively. To evaluate the effectiveness of our method in real observations, we have applied it to a subset of images from the DESI Legacy Imaging Surveys and media images from Euclid Early Release Observations. 61 new strong lensing system candidates are discovered by our method. However, we also identified false positives arising primarily from the simplified galaxy morphology assumptions within the simulation. This underscores the practical limitations of our approach while simultaneously highlighting potential avenues for future improvements.

Figures (16)

• Figure 1: The flowchart of image restoration step, which contains an image restoration neural network (RESTORE) and an image blurring neural network (PSF).
• Figure 2: We chose the r, g, and i bands to generate a color image. Figure (a) shows the raw data, Figure (b) shows the data after the asinh transformation, Figure (c) displays the restoration results and Figure (d) shows the restored image after the asinh grayscale transformation. As depicted in these figures, the incorporation of image restoration and grayscale transformation steps significantly enhances the visibility of image details.
• Figure 3: As depicted in this figure, our method initiates the feature extraction process using the Swin Transformer. These extracted features are subsequently compiled into a feature map, which is then utilized for box regression and classification.
• Figure 4: The specific structure and data flow of the Swin Transformer. The input image undergoes linear embedding, patch merging, Layer Norm, global pooling, and fully connected layers to obtain the final output.
• Figure 5: The diagram illustrates the computation diagram of self-attention with a shifting window. In Layerl, each window is averaged, and the self-attention is calculated independently for each window. In Layerl+1, the windows are shifted systematically, generating new windows that maintain connectivity between the windows when calculating attention.