Research and Application of Automatic Bullet Hole Identification Method
The tests on the bullet hole have an important role in the modern public security and judicial practice, the traditional manual method of visual inspection is time-consuming and low-precision. And the mean absolute difference algorithm based on feature point is proposed. The translation and rotation transformation in the space Cartesian coordinate system are used to build the error analysis model of the bullets image. And the three-dimensional data from the model are revised. In order to process the bad data, the de-noising theory of the wavelet analysis is- used to build the wavelet de-noising model. Using the algorithm based on polygon approximation error for the feature- points to extract the feature points. These feature points are used to identify the bullet hole. The practical application validates the effectiveness and high accuracy of the algorithm.
Error model Wavelet analysis least absolute deviation mean absolute difference algorithm based on feature point
Zhiqiu Li Liang Zhao
Jiangsu Jianzhu Institute, China
国际会议
重庆
英文
553-557
2011-08-20（万方平台首次上网日期，不代表论文的发表时间）