关键词:
光子计数探测器
多能量X射线成像
厚度估算
叠加材料
机器学习
摘要:
自然界中,铜和铁元素常以伴生状态存在于矿石中。为了估算铜铁伴生矿中的元素含量,准确估算铜、铁和石材在叠加状态下的厚度至关重要。本文提出了一种基于多能量X射线透射(ME-XRT)原理的厚度估算方法,利用光子计数探测器(PCD)采集多能量数据,并结合机器学习算法进行系数修正。该方法包括多能量区间的X射线衰减图像获取、图像预处理、特征提取以及基于机器学习的系数修正。实验结果表明,该方法能够有效估算三种材料叠加状态下的厚度分布。In nature, copper and iron elements often coexist in ores. To estimate the elemental content in copper-iron associated ores, accurately determining the thickness of copper, iron, and stone in an overlapping state is crucial. This paper proposes a thickness estimation method based on the principle of Multi-Energy X-ray Transmission (ME-XRT), which utilizes a Photon Counting Detector (PCD) to acquire multi-energy data and combines machine learning algorithms for coefficient correction. The method includes the acquisition of X-ray attenuation images in multiple energy ranges, image preprocessing, feature extraction, and coefficient correction based on machine learning. Experimental results show that the proposed method can effectively estimate the thickness distribution of the three materials in an overlapping state.