The in situ acoustic measurement of seafloor sediment is an important technical means to obtain the acoustic parameters of seafloor. The time‐of‐flight method is commonly used to calculate the sound speed in seafloor sediment. Accurate identification of signal feature points is essential for determining travel time or travel time difference of acoustic signals. However, the precise identification of feature points, such as the take‐off point of the first wave of a sound wave signal, is challenging. The conventional manual identification method is inefficient and prone to errors. The development of a feature point auto‐identification method is imperative for accurately calculating the travel time of acoustic signals. In this study, we employed the cross‐correlation method, the level threshold method and the short window‐long window energy ratio method to extract the acoustic travel time differences and calculate the sound speeds in seawater and in seafloor sediment. We then analysed the effectiveness of these calculated results. The sound speeds in seawater obtained through the aforementioned methods were compared with the sound speeds measured using a sound velocity profiler. The comparison revealed that these processing methods exhibit a high level of accuracy. The sound speed results in sediments show that the programme‐based auto‐identification methods significantly reduce the standard deviation compared to the manual identification method. This study successfully assessed the processing accuracy of different methods and expanded the processing methods for in situ acoustic signals of seafloor sediments.

中文翻译：

---

海底沉积物原位声速测量的时差自动提取方法

海底沉积物原位声学测量是获取海底声学参数的重要技术手段。飞行时间法通常用于计算海底沉积物中的声速。信号特征点的准确识别对于确定声信号的走时或走时差至关重要。然而，特征点的精确识别，例如声波信号第一波的起飞点，具有挑战性。传统的人工识别方法效率低且容易出错。开发特征点自动识别方法对于准确计算声学信号的传播时间至关重要。本研究采用互相关法、电平阈值法和短窗长窗能量比法提取声波走时差并计算海水和海底沉积物中的声速。然后我们分析了这些计算结果的有效性。将通过上述方法获得的海水中的声速与使用声速轮廓仪测量的声速进行比较。比较表明这些处理方法表现出较高的准确性。沉积物中的声速结果表明，与手动识别方法相比，基于程序的自动识别方法显着降低了标准偏差。本研究成功评估了不同方法的处理精度，拓展了海底沉积物原位声信号的处理方法。