Objective To determine the optimal NaCl concentration and evaluation indices for identifying salt tolerance in rice during the germination stage, screen for extreme salt-tolerant and salt-sensitive germplasms, and explore the feasibility of using multispectral imaging technology for the rapid discrimination of seed salt tolerance.

Method Twenty rice germplasms were subjected to a gradient of 10 NaCl concentrations ranging from 0 to 200 mmol/L, and germination data were recorded at 11 time points between 0.5 and 7.0 d of germination. The optimal concentration and evaluation indices were determined by comparing the means and coefficients of variation (CV) of the relative germination rate and relative germination index across the different treatments. Meanwhile, multispectral imaging technology was employed to analyze the spectral reflectance characteristics and normalized canonical discriminant analysis (nCDA) images of extreme germplasms under three conditions: dry seeds, H₂O-imbibed, and NaCl-imbibed.

Result Salt stress exhibited a concentration-dependent inhibitory effect on rice seed germination. Under the 160 mmol/L NaCl treatment, the mean 2.0 d relative germination rate was 0.44 with a CV of 0.55. This condition simultaneously satisfied the criteria for approximately 50% germination inhibition and the maximization of phenotypic differentiation, thereby establishing it as the optimal concentration and core index. The reliability of this evaluation system was validated using an additional 20 germplasms, and based on this, all 40 tested materials were classified into five distinct salt-tolerance levels. Multispectral analysis revealed that under salt stress, the spectral reflectance of salt-tolerant germplasms was significantly higher than that of sensitive ones, exhibiting a clear grouping trend. In nCDA images, salt-tolerant accessions appeared red, while sensitive ones appeared blue, enabling intuitive differentiation.

Conclusion This study established 160 mmol/L NaCl and the 2.0 d relative germination rate as the optimal concentration and core index, respectively, for evaluating salt tolerance during rice germination. Combined with multispectral imaging, this approach enables high-throughput phenotyping of salt tolerance and provides a methodological basis for screening superior germplasms.