With the increasing demand for nuclear energy,uranium exploration has become a key link in the supply of nuclear energy. Uranium exploration methods mainly include radioactive geophysical surveys,geochemical surveys,and other traditional methods,most of which have the shortcomings of inaccurate detection data and low efficiency.This study used near-infrared spectroscopy and stoichiometry to explore the feasibility of screening and identifying uranium super enriched plants. Through the investigation of the growth and characteristics of plants in the uranium mining area,the ultra-enriched plants were selected,the leaf comprehensive spectra in different regions were obtained by a near-infrared spectroscopy analyzer,and the spectral response relationship was compared and analyzed. It was found that the absorption peaks of the two hyper-enrichedplants were located in two bands:650~700 and 950~1 050 nm. The absorption peak of chlorophyll in the former band was mainly generated by the combined frequency of C—O and C—H bond stretching vibration. In the latter band,the absorption peak of water is mainly caused by the 5-order frequency doubling of O—H bond bending vibration. The feature variables were selected by principal component analysis (PCA) and successive projections algorithm (SPA),and the two samples were randomly divided into training and prediction parts according to the ratio of 3∶1,respectively. The detection model of uranium enrichment in super-enriched plants was constructed by combining the two methods of partial least squares(PLS) and least square support vector machine (LSSVM),and the prediction effect was compared. Obtained the detection model of Setaria uranium enrichment based on PLS had the best effect,with a discrimination accuracy of up to 100%,RMSEP of 0.115,and R2 of 0.946. The Setaria detection model is superior to the coverage in the two modeling methods. It may be that the enrichment coefficient of setaria morifera is higher than that of ciderage.The results show that the detection model of uranium enrichment in super-enriched plants established by near-infrared spectroscopy combined with the partial least squares method has the best effect,and it is feasible to screen and identify uranium super-enriched plants. This method provides an important reference for the ecological restoration of the spent uranium ore area and a new idea for the use of specific and indicative plants to search for uranium ore.