摘要
基于高时间效率的FAST算法,提出了一种快速鲁棒的FAST-LOG角点算法。使用直方图均衡化方法对图像进行增强,提高图像成分的清晰度并消除图像中光照强度的影响;运用拉普拉斯-高斯函数对图像进行卷积,实现图像的高斯平滑和增强边缘,及对噪声最大化的抑制;最后使用FAST算子检测角点。对比实验证明,新算子对于添加高斯噪声的分辨率为640*480的图像,其检测时间可达到0.05s;对光照不同的图像具有相近的检测性能;角点重复率可达98%。该算子可应用于实时视频图像的处理,为开发基于视觉的实时智能车辆预警系统提供了新的研究思路。
Based on time-efficient FAST algorithm, the paper described a fast and robust LOG-FAST corner algorithm. Histogram equalization as one kind of image enhancement was firstly applied to the original image for sharpening useful image information and improving image illumination invariance. Then Laplacian of Gaussian operator was convoluted to achieve image guassian smooth and edge enhancement, and also suppress noise in the maximal degree. At last, FAST al- gorithm was applied to produce LOG-FAST corners. The new corner detection algorithm not only features with time-ef- ficient as FAST algorithm, also has illumination invariant, noise invariant and robust feature. Experiments show that LOG-FAST algorithm can achieve a detection time 0. 05s on noised images sized with 640*480. It has the similar detection performance on illuminate variant images and has repeatability 98 percent. Due to its excellent performance, FAST algorithm can be used in real time video process applications such as intelligent vehicle warning system.
出处
《计算机科学》
CSCD
北大核心
2012年第6期251-254,共4页
Computer Science
基金
中国科学院知识创新工程重要方向项目(KGCX2-YW-148)
中国科学院知识创新工程重要方向项目:车联网智能服务应用系统及示范资助