Using uncalibrated digital aerial imagery (DAI) for diagnosing in-season nitrogen (N) deficiencies of corn (Zea mays L.) is challenging because of the dynamic nature of corn growth and the difficulty of obtaining timely imagery. Digital aerial imagery taken later during the growing season is more accurate in identifying areas deficient in N. Even so, the quantitative use of late-season DAI across many fields is still limited because the imagery is not truly calibrated. This study tested whether spectral characteristics of corn canopy derived from uncalibrated late-season DAI could predict corn N status within and across fields. Color and near-infrared  (NIR) imagery was collected in late August or early September across Iowa from 602 corn fields in 2006  and from 690 in 2007. Four sampling areas (one within a target-deficient area as seen in the imagery) were selected within each field for conducting the late-season corn stalk nitrate test (CSNT). The imagery was enhanced to increase the dynamic range and to normalize reflectance values across all fields within a given year. The reflectance values of individual bands and three vegetation indices were used to predict corn N status expressed as Deficient and Sufficient (a combination of  marginal, optimal, and excessive stalk test categories) using a binary logistic regression (BLR). The green reflectance had the largest potential to separate the target-deficient samples from non-deficient samples, and the highest prediction rate in BLR, ranging from 70% for 2006 and 64% for 2007 data. The results confirmed that with the appropriate enhancement method late-season uncalibrated DAI can be used to accurately predict N-deficient and sufficient areas within corn fields.