Based on the stability theory of fractional-order systems, a projective synchronization scheme with different coefficients is realized in [Formula: see text] complex networks to build a model for fractal pattern recognition. In the proposed complex function projection synchronization scheme, first a drive-response network is constructed with [Formula: see text] fractional-order complex nodes; And then reasonable controllers are designed to realize the function projection synchronization with different projection coefficients, which are used to encode the fractal pattern; Finally, using the synchronization error system, the input variables for identifying fractal patterns can be solved, and the fractal pattern can be recognized. From the perspective of neurodynamics, the proposed model provides a better understanding of human perception system.