Abstract

Sediments covering the sea-floor are considered the largest natural archive of paleoclimatic and paleoceanographic information. The study of sediment composition in the coarse fraction (variety of mineral and biogenic grains over 0.063 mm in size) is widely used to pry climate clues out of the sediment record. At present, specialists use a binocular microscope to visually classify grains from a small portion of a sediment sample. This time-consuming technique requires the observer to possess geological expertise. In the previous work, we proposed an algorithm for automatic unsupervised detection of particles and their clustering. In the current work, we present qualitative improvements in the algorithm which now employs the state-of-the-art clustering method, SPICE. This method made it possible to eliminate overclustering and limit the number of clusters to three, making the results more suitable for interpretation. We trained the algorithm and interpreted the obtained results. The resulting model can be used as a classifier, enabling the calculation of particle distribution by clusters, analysis of grain-size distribution, and comparison of these results with those obtained through other lithological analyses. According to the mean deviation from the results of the X-ray diffraction analysis, our method outperforms visual description techniques. Lastly, we developed and deployed an application that automates server-side calculations and allows users to evaluate the of clustering and grain-size measurements.

中文翻译：

---

使用现代深度学习方法从显微摄影中自动估计 >0.1 毫米部分底部沉积物成分的客户端-服务器应用程序

摘要

覆盖海底的沉积物被认为是古气候和古海洋信息的最大自然档案。对粗粒部分（尺寸超过 0.063 毫米的各种矿物和生物颗粒）沉积物成分的研究被广泛用于从沉积物记录中探寻气候线索。目前，专家使用双目显微镜对一小部分沉积物样本中的颗粒进行视觉分类。这种耗时的技术要求观察者具备地质专业知识。在之前的工作中，我们提出了一种自动无监督检测粒子及其聚类的算法。在当前的工作中，我们对算法进行了定性改进，该算法现在采用最先进的聚类方法 SPICE。这种方法可以消除过度聚类并将聚类数量限制为三个，从而使结果更适合解释。我们训练了算法并解释了获得的结果。所得模型可用作分类器，能够按簇计算颗粒分布、分析粒度分布，并将这些结果与通过其他岩性分析获得的结果进行比较。根据 X 射线衍射分析结果的平均偏差，我们的方法优于视觉描述技术。最后，我们开发并部署了一个应用程序，该应用程序可以自动执行服务器端计算，并允许用户评估聚类和粒度测量。