【摘要】 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.更多还原

【Abstract】 In the pendulum single beam bathymeter to detect underwater micro-terrain,the random errors of the elevation data inevitably affect the quality of terrain mapping and terrain prediction accuracy is not measured.Used the characteristics of pendulum single beam detection,an improved median filtering algorithm was proposed.Based on ultrasonic integrated backscatter of the terrain surface characteristics could not be detected,for the first time the introduction of changes in the terrain slope was less than of the experience criteria to determine the smallest jump point interval,the use of the smallest interval to determine whether the measuring point for the jump point.In practice,the use of elevation data of six lines,the results of the standard median filter and the improved median filter showthat the improved median filter algorithm can better protect the terrain feature points.更多还原