Development of an image-based analysis method to determine the physicalcomposition of a mixed waste material

摘要

An experimental study was undertaken to assess the suitability of an image-basedapproach for determining the physical composition of mixed organic wastes.Samples arriving at different waste sorting facilities, each visited twiceduring the study, were physically sorted to examine the composition thesematerials. During these surveys the waste was processed in order to obtaindigital images covering 30 m 2 , representing approximately 250-500 kg of mixedwaste. The images were processed using ERDAS  Imagine software in order toassess the area covered by each component within the waste material. Thecomposition determined from the image analysis was compared with the resultsfrom the physical hand sorting. The image analysis results indicated a strongcorrelation between the physical results (mean r = 0.91) however it was evidentthat components such as film plastics and paper were over-estimated by the imageanalysis approach. This short communication provides initial results,demonstrating the potential  of an image-based method, and discusses furtherresearch requirements and future applications of this technique.