摘要
在航空发动机涡轮叶片的缺陷检测线上,缺陷尺寸的自动测定是实现在线检测的关键。为了解决这一问题,分析了叶片数字射线成像(DR)的纹理特点,通过区域分割点将图像分成不同纹理区,提取各个区域的灰度分布函数以减少纹理对缺陷的干扰。消隐纹理后的图像灰度服从正态分布,遵循背景和信号(缺陷)在正态分布曲线上的取值特点,设置灰度阈值,将灰度值并入集合和数组,最后对航空发动机涡轮叶片的缺陷进行了正确的尺寸和形状测定。
In the on-line testing, measuring the defection automatically is a key-technology. In order to solve this problem about the turbine-vane, the texture of the digital radiography(DR) image was analysed. The image is divided into four regions by means of several intersected points. According to the pixel value in each region, the function expressed the value-distribution is defined and used to decrease the texture in the image. For the pixel value processed and approximating a normal distribution, a threshold is set to get the defection information. Setting function and array are used to describe the defection correctly in the size and the shape at last.
出处
《兵工学报》
EI
CAS
CSCD
北大核心
2005年第3期335-337,共3页
Acta Armamentarii
基金
国防"十五"预研项目 (4 2 0 0 10 5 40 7)
关键词
材料检测与分析技术
数字射线成像
图像纹理
灰度分布函数
正态分布
testing and measuring for material
digital radiography
image-texture
distributed function
normal distribution