Iris segmentation remains a critical yet challenging stage in biometric recognition, especially under offaxis capture, eyelid and eyelash occlusion, specular reflections, and illumination variations that violate the circular and unobstructed assumptions of classical pipelines. We present EDIL-SegRayDP, a training-free and explainable iris segmentation framework that departs from the conventional localization-first paradigm by treating annulus recovery as the primary optimization objective. Rather than committing early to a global center/radius hypothesis and refining it afterward, the proposed method performs segmentation-first boundary recovery with segmentation-aware center rescue and fail-safe outer-boundary control. Occlusion is handled explicitly through geometry-normalized masking and validity-aware annulus construction, while all key parameters are defined in scale-normalized form for cross-dataset portability. Experiments under a fixed-configuration protocol on IITD and CASIA-IrisV4-Interval show strong non-CNN performance with CPU-only inference, achieving an iris-mask mean Dice of 0.9106 on IITD and 0.9377 on CASIA-IrisV4- Interval, with corresponding pupil Dice of 0.9763 and 0.9755. Additional full-benchmark evaluations on CASIA-IrisV4-Lamp and CASIA-IrisV4-Thousand further confirm the portability of the proposed framework across more challenging and larger-scale subsets. Under the evaluation protocol adopted in this study, these results compare favorably with a recent training-free reference, supporting EDIL-SegRayDP as a competitive and interpretable training-free alternative for iris segmentation under non-ideal imaging conditions