In order to measure phase parameters of Sn-Sb-Cu micrographs fast and accurately, a metallographic recogni-tion method based on gray and shape features is given. Determining the inclusion relationship of polygons by improving the test method of intersection-point count can greatly reduce the probability of graphics misrecognition. Compactness, an extracted feature, can describe difference of each phase better. Experimental results show that the proposed method has a high recognition rate for metallographic, the obtained relative amounts of each phase are close to measured data of XRD spectra, and using the extracted phase parameters can establish a quantitative relationship between the organization and hardness.%针对Sn-Sb-Cu显微图中各相参数的快速、精确测量,给出一种基于灰度和形状特征的金相联合识别方法.改进的交点计数检验法判断多边形的包含关系能够降低图形误识别的概率,提取的紧凑度这一特征可较好地反映各相差异.实验结果表明该算法对金相具有较高的识别率,获得的各相相对量与XRD能谱测得的数据较为接近,利用提取的各相特征参数建立了组织和合金硬度的定量关系.
展开▼