Heavy metal ions in plants can be determined by using the near-infrared spectral (NIRS) technique, because they combine with the organic molecular groups that have NIRS absorptions. The present article analyzed the fast detection of heavy metal Cu in Ludwigia prostrata leaves by near infrared diffuse spectral technology. Different preprocessing methods were compared, combined with partial least squares (PLS), and the fast detection models of heavy metal Cu in Ludwigia prostrata leaves were established. The results showed that the best model was obtained by PLS with the preprocessing method of average smoothing. The correlation coefficient (r) and root mean square error of calibration(RMSECV) was 0.950 and 5.99 respectively; External validation correlation coefficient (r) and root mean square error of prediction(RMSEP) was 0.923 and 7.38 respectively. The study shows that fast determination of heavy metal Cu in Ludwigia prostrata leaves using near infrared diffuse spectroscopy is feasible.