无人机航拍视频码率控制算法优化

OPTIMIZATION OF RATE CONTROL ALGORITHM FOR UAV AERIAL VIDEO

高效率视频编码(HEVC)在实际应用中很难满足无人机航拍领域的需求。针对此问题,提出一种基于缓冲区状态及视频显著性的码率控制算法。对预处理后的帧进行颜色、纹理特征提取,并采用log-Gabor滤波器对其进行特征滤波,得到当前帧的显著图;计算每帧和每个最小单元(CTU)的显著性权重;以帧的显著性权重和缓冲区状态指导帧层比特分配,以CTU的显著性权重指导CTU层比特分配。实验结果表明,提出的算法可以在缓冲区维持在一个平稳状态的前提下,提高0.61 dB视频显著区域的峰值信噪比,更适用于无人机航拍的场景。

High efficiency video coding(HEVC)is difficult to meet the requirements of UAV aerial photography in practical applications.Aimed at this problem,a rate control algorithm based on the buffer status and video saliency is proposed.Color and texture features were extracted from the preprocessed frames,and the log-Gabor filter was used to filter them to obtain the saliency map of the current frame.The saliency weight of each frame and each CTU were calculated.The saliency weight of the frame and the buffer status were used to guide the frame layer bit allocation,and the saliency weight of each CTU were used to guide the CTU layer bit allocation.Experimental results show that the proposed algorithm can maintain the buffer in a stable state,improve the peak signal-to-noise ratio of 0.61 dB in video significant area,which is more suitable for drone aerial photography applications.