Imaging Fluorescence Measurements of Black Plastic Particles Measured With 450 nm Excitation and a Demonstrator System

The data contains imaging fluorescence measurements of black plastic particles in the spectral range from 500 nm to 900 nm. The files are stored in the envi-format used typically for hyperspectral data. Therefore there are two files for each measurement: the data (.envi) and the header-file (.envi.hdr). The files can be opened with software like ENVI or the Imanto<i>viewer </i>(https://www.iws.fraunhofer.de/de/geschaeftsfelder/chemische_oberflaechentechnik/prozess_monitoring/produkte/imanto.html). The measurements were made with a demonstrator system developed for imaging fluorescence measurements in an industrial like environment. For excitation four 4000 mW 450 nm diode laser covering a length of 40 cm were used. The samples were also illuminated with a 870 nm LED. For each polymer several thousand particles were measured. For the measurements the plastic particles were placed on a conveyer belt and measured with the hyperspectral scanning system. Three measurements were carried out per polymer.The measured polymers are HDPE, PP, SBR, PA6, PA6 with glassfiber, PA66, PA66 with glass fibre, PA66 with flame retarder, TPU, TPE, TEEE, PPS, PBT and POM. The samples were obtained from a recycling plant and the plastic was grinded cryogenically at the plant.The data was used to train classification models. The objective is to use the method for recycling black plastics.<br><br>

本数据集包含500 nm至900 nm光谱范围内的黑色塑料颗粒成像荧光测量数据。数据采用高光谱数据常用的ENVI格式（envi-format）存储，因此每次测量对应两个文件：数据文件（.envi）与头文件（.envi.hdr）。此类文件可通过ENVI或Imanto Viewer（https://www.iws.fraunhofer.de/de/geschaeftsfelder/chemische_oberflaechentechnik/prozess_monitoring/produkte/imanto.html）等软件打开。本次测量采用专为类工业环境下成像荧光测量开发的演示系统完成，激发光源采用4台功率4000 mW、波长450 nm的二极管激光器，覆盖长度达40 cm，同时采用870 nm LED对样品进行照明。每种聚合物均测量了数千颗颗粒，测量时塑料颗粒被放置于传送带上，通过高光谱扫描系统完成采集，且每种聚合物完成三次测量。本次测量涉及的聚合物包括HDPE、PP、SBR、PA6、玻纤增强PA6、PA66、玻纤增强PA66、阻燃PA66、TPU、TPE、TEEE、PPS、PBT以及POM。样品取自某回收工厂，塑料颗粒在该厂通过低温工艺研磨制备。本数据集可用于训练分类模型，其应用目标为实现黑色塑料的回收分选。