According to the measurement requirements of micro-drill, a new approach based on hybrid mutation neural network integrated with fuzzy adaptive particle-swarm optimization (PSO) for fitting of micro-drill′s feature curves is presented. The network is designed to coincide with the fitted equation in experiment. After the training of network and every iteration of particle individual, the obtained neural weights are normalized to form a unit weight vector, which is equivalent to a special mutation operation for individuals. At the same time, the fuzzy adaptive PSO is integrated into the solving algorithm to get the global optimization solution. And the inertia factor of PSO is tuned self-adaptively by adopting fuzzy logic reasoning according to the characteristic of particle′s motion trajectory in longitudinal direction and lateral direction. When the solving system comes to the global equilibrium point, the position vector of the best particle is used to obtain the expression coefficients of the fitted equation. Then in the light of the fitting equations, the structural parameters and flank faults such as core width, rounded corners, chips of the micro-drill and so on can be obtained easily. Compared with the traditional test approaches such as least square method, experimental results show that the proposed approach provides a new scheme for the curves fitting of micro-drill and other work-pieces with high measurement precision.