The actual application environment of airborne radar presents non-uniform characteristics, resulting in poor clutter suppression effect and high false detection rate of high-altitude weak moving targets. Therefore, research is being conducted on high-altitude weak moving target detection methods for airborne radar in non-uniform environments. Firstly, the Doppler compensation method is used to improve and optimize the traditional space-time adaptive processing algorithm (STAP), and based on the improved STAP algorithm, non-uniform environmental clutter in radar echo signals is suppressed. Then, based on a bandpass filter, the radar echo signal is filtered to enhance the target signal and extract the micro Doppler features of the radar echo signal. Finally, based on the characteristic vector of the radar echo signal of high-altitude weak moving targets, a high-altitude weak moving target detection model is constructed using support vector machine (SVM), and the detection threshold is adaptively set to obtain accurate high-altitude weak moving target detection results. The experimental results show that after applying the method proposed in this paper, the maximum amplitude of non-uniform environmental clutter is only 0.04dB, effectively reducing the interference of clutter on target detection; The time-frequency spectrum results are consistent with the actual time-frequency spectrum results, verifying the accuracy of the proposed method for signal feature extraction; The minimum false detection rate of high-altitude weak moving targets reaches 0.1%, reducing the probability of false detection and providing an effective solution for accurate detection of high-altitude weak moving targets by airborne radar in non-uniform environments.