Phytoene dehydrogenase (PDS) and p-hydroxyphenylpyruvate dioxygenase (HPPD) are two important herbicides targets in photosynthesis. If inhibiting the activity of double target enzymes, the synthesis of carotenoids and plastoquinones is blocked and leads to albinism and death of weeds. HPPD and PDS dual-target pharmacophores based on molecular common features were constructed to obtain the intersected compounds through virtual screening. Compound with good predict activity (ZINC000511042115) was imported into AILDE for optimization to generate 15 new inhibitors, and molecular docking was performed to identify the interaction with key residues in the active site. The docking results of 15 compounds were all better than the native ligand in HPPD protein and were at the same level as norflurazon in PDS protein. Five possible dual-target inhibitors were obtained by virtual screening and activity prediction. The present work provides powerful insights into the development of novel dual-target inhibitor herbicides with computational techniques.