论文部分内容阅读
在摆动式超声单波束探测水下微地形中,高程数据的随机误差不可避免地影响地形绘制质量及未测地形预测精度.结合摆动式单波束探测特点,提出了一种改进型中值滤波算法,即根据背向超声波的地形坡面无法被探测的特点,首次引入地形坡度变化小于的经验准则确定跃变点最小间隔,再利用最小间隔判断测点是否为跃变点.在实际应用中,利用6条测线的高程数据,由标准中值滤波与改进型中值滤波进行比较.结果表明:改进型中值滤波算法能更好地保护地形特征点.
In underwater micro-topography with wobble ultrasonic single-beam detection, random error of elevation data will inevitably affect the quality of terrain rendering and the accuracy of terrain prediction.An improved median filtering algorithm , That is, according to the characteristics that the terrain slope can not be detected by the ultrasonic wave, the minimum interval of the transition point is determined by the experience criterion that the variation of the terrain gradient is less than the first, and then the minimum interval is used to judge whether the measurement point is the transition point.In practice, The standard median filter and the improved median filter were compared by using the height data of six lines.The results show that the improved median filter algorithm can better protect the topographic feature points.