In the wear and tear process of synovial joints, wear particles generated and released from articular cartilage within the joints have surface topography and mechanical property which can be used to reveal wear conditions. Three-dimensional (3D) particle images acquired using laser scanning confocal microscopy (LSCM) contain appropriate surface information for quantitatively characterizing the surface morphology and changes to seek a further understanding of the wear process and wear features. This paper presents a new attempt on the 3D numerical characterisation of wear particle surfaces using the field and feature parameter sets which are defined in ISO/FDIS 25178-2. Based on the innovative pattern recognition capability, the feature parameters are, for the first time, employed for quantitative analysis of wear debris surface textures. Through performing parameter classification, ANOVA analysis and correlation analysis, typical changing trends of the surface transformation of the wear particles along with the severity of wear conditions and osteoarthritis (OA) have been observed. Moreover, the feature parameters have shown a significant sensitivity with the wear particle surfaces texture evolution under OA development. A correlation analysis of the numerical analysis results of cartilage surface texture variations and that of their wear particles has been conducted in this study. Key surface descriptors have been determined. Further research is needed to verify the above outcomes using clinic samples.