枪弹头痕迹自动比对方法的研究

Research on Automatic Matching Methods of Bullet Marks

对枪弹头痕迹自动比对方法问题进行建模分析，结论如下：1）提出了姿态标识线的概念，将姿态标识线的斜率和截距作为标识弹头位置和姿态的两个指标以此来消除由人工测量造成的误差．2）通过对原始信息进行去噪、变化坐标等预处理，绘制等高线图，提取弹痕几何对比特征；通过显著性检验确定相关系数大于0．4038的具有显著相关关系区域作为一个稳定且相关度高的弹痕测量数据区域；3）对任意两个弹头分别计算不同比对偏移量时各个次棱所各自对应的相关系数和误差矩阵，寻找最佳次棱对比，然后对比对过的两个弹头计算两者之间的整体相似程度，列表给出两两之间的相似程度．

The automatic matching methods of bullet marks are modeled mathematically in this essay. The results are as follows： 1） the concept of pose marking line is brought forward, and the slope and intercept of the pose marking line was used as two indexes marking the pose and position of the bullet to reduce the errors of the manual measuring; 2） by removing the noise and changing coordinates on the original information, the contour map was plotted, and the comparing characteristics of the bullet marks geometry was extracted; by testing the significance correlation coefficients of the region in which the coefficients is greater than 0.4038 is thought as a stable and high relevant bullet-ridden measurement data region; 3） by calculating the correlation coefficients and error matrix according to each sub edge in different offsets of the matching for two random warheads, the best sub edge contrast was found, then the degree of similarity between the two was listed by comparing the overall similarity of the contrasted warheads.