The major objective of this paper is to establish a drone’s appearance design evaluation model based on BP neural network, so as to make the drones appearance design more scientific and meet the actual needs. Evaluation criteria for drones' visual styles using design based neural networks is to establish a comprehensive evaluation index system, organize experts to evaluate and score the samples, obtain the training and inspection data of neural network, assess the relative significance of each index by MATLAB simulation, and extract the nonlinear relationship reflecting each metric used in evaluating, and the overall assessment's findings. Model for testing BP neural networks reduces the effect of subjective factors in the evaluation process of comprehensive assessment method and analytic hierarchy process, can objectively and comprehensively evaluate the drone’s appearance design, and provides an effective reference for selecting the optimal drones appearance design scheme.