摘要
全自动金丝球焊机是半导体制造后端工序的关键设备之一,本文提出应用同态增晰滤波法增强图像的对比度,减少了光照变化并锐化了边缘和细节图像,去除了因光照不均而引起图像像素灰度的低动态范围,然后再用多图像平均法以降低信噪比,完成芯片图像的预处理,为全自动金丝球焊机芯片图像的定位奠定了基础。
The full automatic gold wire bonder is one of the key equipment of semiconductor manufacturing's backend processing. This article puts forward the method which named homonorphic norm of the filter to enhance the contrast of the image, and reduce the illumination change as well sharpen the image edges and details, removing the low dynamic range of causing image pixel which is because of uneven illumination, and then with the average of some images method to get rid of the noise lower signal-to-noise rations, it accomplished the preprocessing of the chips, this method laid a solid foundation for the full automatic gold wire bonder chip image positioning.
出处
《中国科技信息》
2014年第2期114-116,共3页
China Science and Technology Information
关键词
同态增晰滤波
多图像平均法
预处理
金丝球焊机
homonorphic norm of the filter
the average of some images method
preprocessing
goldwire bonder