The internal logic of the three phase stator current’s Park vector locus and the fault features of a motor were analyzed, and the Park vector locus′ pixel can be used to diagnose the stator winding faults of induction motors. Then, a method diagnosing the stator turn-to-turn short circuit faults based on Park vector locus′ pixel coding in polar coordinates was presented as viewed from pattern recognition. The fault feature vector can be extracted by the pixel matrix code, and the faults were reco gnized by the RBF neural network automatically. Experimental results demonstrate that the method can be used to diagnose the induction motors stator faults, and the precision of the faults recognition is improved.