Abstract: The airborne imager of cloud particles is a key scientific instrument that can directly obtain the properties of size distributions and images in clouds. It plays an important role in probing cloud microphysical structures, verifying remote sensing data, improving cloud microphysical processes in numerical weather models and operating weather modification. Therefore, it is important to continuously improve and develop the instrument of higher probing ability. This study introduces the principles of measurement, the main performance indexes, the calibration method and the application results of observational experiments using the airborne cloud imager that was newly developed in 2011. By repeated trials for several years, some key technologies such as beam homogenization, weak signal detection, multiplex parallel processing algorithm and particle diffraction elimination have been either solved or greatly improved. The homogenization and consistence of optical power have been greatly increased. The mean optical power density in light spot is increased by about three times, and thus the imaging ability has been greatly improved. The higher resolution of linear photo diode array is used with high response speed, which apparently increases the detection ability for small-size cloud particles. A series of basic tests of parameters, system optimization and environmental adaptation, and over 60 in-situ flights have been conducted during the period of study and development of the instrument. Compared with observations of an imported instrument in two flights on 5 November 2018, the preliminary experiment results using the newly development instrument indicate that the accuracy of number concentrations for small-size particles can increase up to one order of magnitude. Meantime, the instrument has higher stability and reliability during the flight, and can correctly obtain clearer cloud habits.