The developed algorithm was used for the real time specific weed discrimination employing multi-level wavelet decomposition. This algorithm used four different types of wavelets i.e., Daubechies (bd4), Symlets (sym4), Biorthogonal (bior3.3) and Reverse Biorthogonal (rbio3.3) up to four levels of decomposition to classify images into broad and narrow class for real-time selective herbicide application using the Euclidian distance method. The lab, which have shown that the system to be very effective in weed identification, segmentation and discrimination. The test and analysis show that 97.26 classification accuracy over 350 sample images (broad & narrow) with 175 samples from each category of weeds and the proposed algorithm takes 29 ms as average time for the classification of the specific weeds.