Soybean pod-related traits are important for seed development, yield formation, and cultivar evaluation. However, conventional measurements are inefficient and have limited ability to characterize complex pod features such as curvature, local enlargement, and continuous color variation. In this study, mature pod images of 187 cultivated soybean accessions collected across two successive years were analyzed using a deep learning-assisted image phenotyping approach. Eight pod-related traits related to size, morphology, and color were extracted from pod images. A genome-wide association study (GWAS) was performed using 61,541 high-quality SNP markers to dissect the genetic basis of these image-derived pod traits. A total of 16 stable loci associated with pod size, morphology, and color traits were identified across 11 chromosomes. Among these loci, eight were not reported in the previous image-based soybean pod GWAS study. Based on SoyBase gene annotation and Gene Ontology biological process information, 32 biologically relevant candidate gene records were prioritized within the corresponding candidate genomic intervals, while pod-related expression profiles and SoyBase association information were used as supporting evidence for candidate gene evaluation. These findings indicate that refined image-derived traits can provide complementary genetic information beyond conventional pod measurements and offer additional opportunities for dissecting soybean pod development, morphology, and mature pod color variation.

Soybean pod-related traits are important agronomic characteristics associated with seed development, domestication, cultivar identification, and breeding improvement. However, conventional phenotyping methods mainly rely on manual measurements, which are time-consuming and labor-intensive and capture only limited dimensions of pod variation, while the genetic basis of skeleton- and curvature-based pod descriptors remains insufficiently characterized in biparental populations. In this study, eight quantitative traits representing pod size, shape, and color components were extracted from an existing mature pod image dataset of an interspecific soybean recombinant inbred line (RIL) population using the established deep learning-based image phenotyping framework. These traits exhibited substantial phenotypic variation, with across-year entry-mean broad-sense heritability (H²) estimates ranging from 0.30 to 0.80. Composite interval mapping (CIM) based on a high-density genetic linkage map identified 54 quantitative trait loci (QTLs), which were integrated into 39 non-redundant loci, including six cross-year stable QTLs and three QTLs supported by best linear unbiased prediction (BLUP) analysis. Candidate genes within selected focal QTL regions were prioritized through functional annotation and pod and seed developmental expression analyses. Among them, Glyma.17G109100 (GmSW17) was prioritized as a positional candidate gene for pod size-related variation, whereas Glyma.19G120400 (L1), a previously validated causal gene for pod color, was located within qV19. These findings demonstrate the effectiveness of combining deep learning-based phenotyping with genetic analysis for dissecting the genetic architecture of complex soybean pod-related traits and provide valuable stable QTLs and candidate genes for future functional studies and soybean molecular breeding.