In view of the problems in the traditional aircraft detection algorithms and the existing machine learning detection algorithms, one concept of valid aircraft detection is proposed for remote sensing images. With the full convolution detection and segmentation network based on the deep learning in the cognitive model, one valid aircraft detection system is designed and simulated. A cognitive model for detection is constructed, and the function of each module is designed. The experimental results certify the effectiveness of this system, and this system provides a new thinking way and method for the development of intelligent detection of multiple objectives.