The introduction of transgenic glyphosate-tolerant crops has led to the evolution of many resistant weeds due to the increased use of herbicides, particularly during the post-emergent growth of crops. The fixation of superior plant phenotypes is an important stage in the molecular plant breeding programme.A robust genotype-phenotype relationship is essential for breeding desirable big bluestem with high fermentable sugar contents, although conventional wet chemical approaches are expensive and time-consuming. Here, non-transgenic maize (non-GM) and transgenic maize (GM) were studied at 2, 4, 6, and 8 days after the beginning of the glyphosate treatment. Chlorophyll a fluorescence image and hyperspectral image technique were applied to monitor the performance of plants and analyse their protective responses. Specific physiological effects of glyphosate affected the photochemical processes of maize cultivars, which induced substantial changes in fluorescence induction characteristics. Our result indicated that F_v and F_q were sensitive to glyphosate and could be used as fluorescence phenotype markers. Simultaneously, hyperspectral image were also applied to evaluate the performance of different cultivars. The measurement of shikimic acid accumulation in response to glyphosate is a rapid and accurate assay to quantify glyphosate-induced damage in sensitive plants. Hence, the combination of spectroscopy and chemometric methods was implemented to predict shikimic acid content. Our result indicated that partial least-squares regression (PLSR) model built on optimal wavelengths had a satisfactory result for predicting shikimic acid contents with correlation coefficient value of 0.89 for calibration set, and 0.78 for predication set. Moreover, shikimic acid concentration in different plant were visualized on the predication maps by prediction spectral features on each hyperspectral image pixel. The overall results of this study indicated that the chlorophyll fluorescence imaging technique and hyperspectral imaging technique could both provide a useful tool for phenotyping in maize breeding programs and could enable the detection and elimination of unwanted genotypes at the beginning of the selection process.