Responses of the middle and upper atmospheric (80-100 km ) wind to geomagnetic activities have been investigated using neutral wind data from 2012 to 2018 years, which were observed by Mohe, Beijing and Wuhan Meteor radars. Daily averaged wind data for geomagnetic quiet condition (Kp ≤ 2) and geomagnetic disturb condition (Kp ≥ 4) were chosen for comparison, and the variation characteristics of wind during geomagnetic disturbances were obtained. The observations show that the influence of geomagnetic activity on zonal wind varied with seasons and latitudes. For zonal wind, the effect of geomagnetic activity at higher latitudes tended to be more westerly wind in the upper mesosphere and more easterly wind in the lower thermosphere, and the differences between disturbed and quiet conditions were on the order of 3 m/s; while for the lower latitudes, it tended to be more easterly wind in the 80-100 km region, and the influence were about 5 m/s. In spring, the three stations had similar tendencies, and had no latitude differences. But the easterly wind in the middle atmosphere became stronger with the decrease of latitude in summer/winter. The effect of geomagnetic activities on the meridional wind had seasonal differences. The influence of geomagnetic activities in spring and winter was stronger than that in summer and autumn. In winter, the effect of geomagnetic activity on the meridional wind in middle and low latitudes was stronger than that in higher latitudes. According to the calculation results, the influence on zonal wind was about 5 m/s to 10 m/s, and on meridional wind was about 3 m/s to 5 m/s. The impact of geomagnetic activities on MLT wind can penetrate down to about 80 km. At this height, the influence on zonal wind was the strongest in spring, reaching 8 m/s, and on meridional wind was the strongest in spring/winter, reaching 5 m/s.

The prerequisite of target detection and tracking is to model and represent the target based on multi-dimensional features extracted from the target region. The traditional target feature extraction needs to connect the target region first and then must calculate the target feature, which has still room for improvement in real-time. An advantage of the new method based on the synchronous calculation of pixel-connected domain markers and target features is that the target features can be output when the target region is connected. The process establishes a feature transfer mechanism, creates a marker table, a marker mapping table, and a feature attribute table simultaneously when scanning images, and uses the marker mapping table to associate the marker table and the feature attribute table. The marker merging and feature attribute transfer calculation are performed synchronously at the time of region adjacency, which ensures the real-time target feature extraction. As a result, the method proved to be effective because of an implementation scheme based on an FPGA system design. Simulation test results show that this method has several outstanding features: the time of connected domain markers is close to the theoretical minimum; the storage of images utilizes the circular buffer with low resource consumption; marking and computation are processed in parallel flow to improve detection and tracking efficiency; multi-target features are tested and verified to be accurate, which can effectively support subsequent target tracking detection; and it has theoretical and practical values.