Impact damage has a significant impact on the safe operation of composite structures， and there is an urgent need to develop structural health monitoring technology to detect impact damage in a timely manner. This paper proposes a method for preparing carbon nanofiber films and co-curing them on the surface of composite laminates. The distribution image of surface conductivity changes caused by impact damage is reconstructed using electrical resistance tomography technology to monitor and identify impact damage. Tikhonov regularization and SpaRSA sparse regularization methods were respectively used to reconstruct the damage image. Impact tests and damage identification tests were conducted on composite laminates， and the results have showed that the combination of electrical resistance tomography and carbon nanofiber film is effective for impact damage identification； both Tikhonov regularization and SpaRSA sparse regularization methods can reconstruct the electrical conductivity changes of carbon nanofiber film layers caused by damage. The difference between the two methods in identifying the location of the damage center is small， but both methods have certain shortcomings in identifying the size of the damage. Integrating the identification results of the two methods through image fusion can improve the accuracy of damage identification.