基于曲线拟合的双能X射线透射的金属识别研究

Metal Recognition Algorithm Based on Curve Fitting for Dual-Energy X-ray Transmission

双能X射线透射分析技术通过识别物料种类来回收废旧金属。针对传统的识别算法存在识别结果受金属厚度影响大、可识别的金属厚度范围较小的问题,在曲线拟合方法的基础上提出了一种基于野值判断的金属识别方法。采用基于高、低能T值的拟合曲线,减少了厚度因素对拟合曲线相关度的影响,解决了较厚金属无法识别的问题;引入野值判断,提高了识别的准确性。对不同厚度的金属铝和金属铜进行识别实验,结果证明:基于野值判断的T值拟合识别算法对不同厚度铝和铜的识别效果较好,可识别的厚度范围相较于传统算法分别提高至90 mm和40 mm,扩大了可以分选的金属厚度范围。

Dual-energy X-ray transmission analysis technology recycles scrap metal by identifying material types.In order to solve the problems that the identification result is greatly affected by the metal thickness and the range of metal thickness can be recognized is small,a metal identification method based on outliers is proposed based on the curve fitting method.The fitting curves based on high and low energy T-values are used to reduce the influence of thickness factor on the correlation of fitting curves,and the problem that the thick metal cannot be identified is solved.Outliers judgement is introduced to improve the accuracy of recognition.The identification experiments of metal aluminum and metal copper with different thicknesses are carried out.The results show that the T-value fitting identification algorithm based on outliers judgment has a better identification effect on aluminum and copper with different thicknesses.Compared with the traditional algorithm,the recognizable thickness range is increased to 90 mm and 40 mm respectively,which expands the range of metal thickness that can be sorted.