Event cameras are increasingly popular in the field of artificial vision systems as a new type of biological vision sensor. Different from traditional detectors, which capture images at a fixed rate, event cameras are inspired by biological retinas and respond asynchronously and independently to changes in the brightness of each pixel in the scene. The pixel unit independently detects the change range of the logarithmic value of the light intensity. When it exceeds the preset threshold, the pixel unit is activated and reads out the relevant information of this activation through an external unit, generally including the position of the pixel, activation time, polarity, and other information. When read out, these information are packaged into a data packet, which is called an "event". After the event is output, the photosensitive pixel unit will return to the inactive state and re-monitor the changes in the outside tight intensity. This working principle makes event cameras offer attractive properties compared to traditional cameras: large dynamic range (140 dB vs. 60 dB), high temporal resolution (in the order of us), low latency, low power consumption, and high pixel bandwidth ( in the order of kHz) resulting in reduced motion blur. Hence, Event cameras are widely used in spatial vision problems such as pose estimation, 3D reconstruction, and SLAM. With the development of the event camera, it has been gradually applied in the field of space target detection, and has shown great potential. Event cameras provide a new solution for monitoring satellites and stars for space situational awareness. However, there are few research on space situational awareness based on event cameras. Aiming at the problem of space target detection based on event cameras, we clarify the mechanism of event camera for space target observation, and systematically analyze the influencing factors that affect the detection sensitivity of space target based on event camera. Apart from theoretical analysis, we also build a detection sensitivity model of space target based on event cameras. As described by the proposed sensitivity model, the optical system parameters of the observation system, such as the clear aperture of the observation optical system, the diameter of the dispersion spot caused by the optical system, the transmittance of the optical system, the photoelectric conversion capability of the camera, and the atmospheric turbulence, determine the limit observation capability of the system. As the threshold of the event camera increases, the detection sensitivity of the system (characterized by limit magnitude) decreases linearly. Besides, we performed field experiments with telescopes using the CMOS camera and event camera. First, we conducted an experiment to observe the planetary based on the event camera. The observation results show that the event camera can be applied to space target observation. Compared with the traditional CMOS camera, the event camera has a lower spatial resolution, hut the characteristics of its low bandwidth communication, low weight, low power, and high speed make it fully meet the harsh requirements of sensors in the aerospace field, and have great application prospects in space target monitoring. In order to explore the impact of the event camera threshold setting on the detection results, we observed Polaris at different thresholds, and the results show that the low threshold makes the signal in the image more obvious but more noise events. When the weak target is observed, the too-low threshold is easy to cause the indistinguishability of noise events and signal events. When the threshold is raised, the detection sensitivity of the system decreases, and the signal is not obvious, which affects the signal recognition. Therefore, when using the event camera for space target observation, the selection of the threshold value is extremely important, riot only to ensure high detection sensitivity, but also to avoid the generation of a large number of noise events, which will affect the signal recognition. So establishing a sensitivity model of space target detection based on event camera can provide theoretical guidance for experiments. For the observation system we built, we selected five celestial objects that are easier to distinguish in the sky to calibrate the sensitivity model. The fitting results show good linearity with correlation coefficient greater than 0.95. After that, we chose three other celestial bodies to verify the model. The relative error was less than 3%, so the model we built is accurate. Event cameras are revolutionary sensors that offer many advantages over traditional, frame-based cameras. Their characteristics, such as low latency, low power, high speed and high dynamic range, make event cameras have a large potential for space target detection. In the meantime, many challenges remain ahead. By conducting theoretical analysis and experimental verification, this paper obtains the sensitivity model of space target detection based on event camera. We hope it can provide a few theoretical guidance for space target observation and other related research based on event camera.
