In a complex environment, it is very difficult to detect the target, and the target is easily affected by human beings or environment. In this paper, the algorithm Faster RCNN convolution neural network (FAHL‐Convolution Neural Network) based on deep learning is adopted by transfer learning method to solve this problem. In military activities, tank targets are camouflaged and hidden in the jungle. In order to improve detection accuracy and reduce the rate of missed detection, Faster RCNN algorithm is combined with RPN network to achieve the effect of feature fusion, and the candidate areas are optimized, which can not only improve the detection accuracy but also ensure the speed. Experimental results show that the improved algorithm has higher detection accuracy and robustness.