基于广义直方图均衡的图像增强新方法  

A Novel Method for Image Enhancement Based on Generalized Histogram

作　　者：刘松涛[1,2] 吴钢[2] 

机构地区：[1]大连理工大学,辽宁大连116024 [2]海军大连舰艇学院,辽宁大连116018

出　　处：《电光与控制》2010年第3期12-14,39,共4页Electronics Optics & Control

基　　金：国家自然科学基金资助(60572160);学院科研发展基金资助

摘　　要：针对传统直方图均衡方法进行图像增强时导致图像细节信息丢失和噪声放大的问题,提出了广义直方图均衡图像增强方法。传统直方图是基于整数1进行计数的,广义直方图将整数1分解成2个值:分数和余数,通过对每个灰度级累计分数值和在所有灰度级均匀分配余数值生成。由于分数值控制着广义直方图均衡的效果,图像中不同特性区域应该具有不同的分数值,给出了基于局部复杂度计算的分数值求取方法。实验结果表明本方法能有效克服传统直方图均衡法丢失目标细节和过分放大噪声的不足,是一种有效的图像增强方法。In order to solve the problems of image detail loss and noise magnification when using the traditional histogram equalization method, a novel method for image enhancement based on generalized histogram is proposed. In the traditional histogram the integer 1 is allocated to a pixel, but in the generalized histogram it is decomposed into two values of the fractional count and the remainder count, then the generalized histogram is generated by accumulating the fractional count for each intensity level and distributing the remainder count uniformly throughout the intensity levels. Since the effect of generalized histogram equalization depends on the fractional count, different image region should have different fractional count, thus the fractional count is given by computing local spatial complexity. The experimental results show that the proposed method can overcome the common histogram equalization method＇s shortcomings, and is an effective image enhancement method.

关 键 词：图像增强 直方图均衡 广义直方图 局部复杂度 

分 类 号：V243.6[航空宇航科学与技术—飞行器设计]