铁谱技术是机车内燃发动机磨损工况监测与故障诊断的重要油液分析手段之一，本文将其信息处理过程中的关键技术难点和重点——“磨粒图像自动识别与分析”作为主要研究对象，在分析综述近年有关研究方法的基础上，提出了基于神经网络解决彩色磨粒图像自动识别问题的思路、方法以及相应的处理流程、模型和软件，并给出有关实验和初步应用结果，论证了这一方法的可行性、应用价值与前景。该流程可用于其它微粒图像（如血细胞和显微金相组织等）自动识别领域 Ferrography is one of the most important oil analysis methods for the monitoring and fault diagnosis of locomotive internal combustion engine wear. In this paper, the key technical difficulties and key points in the process of information processing - “automatic identification and analysis of abrasive grains” Based on the review of the related research methods in recent years, this paper puts forward the idea, method, corresponding processing flow, model and software based on neural network to solve the problem of automatic identification of color abrasive grains. The results of experiment and preliminary application are given, The feasibility, application value and prospect of this method are demonstrated. The process can be used in other particle images (such as blood cells and microstructure, etc.) automatic identification field