Taking unmanned aerial vehicle (UAV) swarm as an example, this paper analyzes the detectability of an air-space infrared detection system for the UAV swarm. Firstly, the UAV swarm flight process is divided into three phases, and its infrared radiation characteristics and imaging characteristics are modeled and analyzed in detail. Secondly, as dispersion characteristics of target imaging in the detection range modeling are considered inadequately, a noise equivalent flux density (NEFD) point target detection range model based on dispersion coefficient is established. In the meantime, in order to verify the effectiveness of the proposed model, different modeling methods and dispersion coefficient comparison methods are used to analyze the change of detection range with signal noise ratio (SNR) and the swarm speed. Simulation results show that the detection range of the proposed model is better than that of other models, and the detection range increases as the swarm flight speed and SNR increase. Furthermore, the dispersion coefficient varies with the detection range. The detection range obtained by the dispersion coefficient solution is superior to that obtained by other methods.