摘要
针对Bernsen算法的缺陷,提出一种名为BM(Bernsen and Mean)的新算法,该算法对于细节较多、形状多样的图像有着良好的处理效果,且能够克服不均匀光照的影响。通过调节算法中的参数就能够满足不同场合和不同类图像的需要。用其得到的目标图形完整真实,且噪声点很少。利用细节信息丰富的车标图片对BM算法进行测试,结果表明该算法整体效果优于其他算法。
Aiming at the defects of Bernsen algorithm, this paper proposes a new algorithm called BM（Bernsen and Mean）. The algorithm has a great processing effect for the image which contain rich details and a variety of shapes. It can also overcome the influence of uneven illumination. The algorithm can satisfy the needs of different occasions and images by adjusting the parameters in the algorithm. It can get target figure integrated with little noise. Use auto logo which have rich details to test BM algorithm, the overall result is superior to other algorithms.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2014年第8期127-132,共6页
Journal of Wuhan University of Technology
基金
国家自然科学基金(61303028)
