摘要
针对传统自动曝光算法不能适应背光或正面强光等复杂光照环境,且在实时性和平滑性不能同时兼顾等方面的缺点,提出一种基于图像多区域的变步长自动曝光算法。该算法将主体区域和背景区域进行分区,并对主体区域赋予更大的调整权重值来计算图像平均亮度,最后根据平均亮度距离目标亮度的差值等级来调整曝光时间的变化步长。实验结果表明,利用这种曝光方法在复杂光照环境下图像主体区域也能得到很好地体现,且在光照强度不断地变化时,图像亮度也能快速平稳地接近理想亮度,具有很好的成像效果。
Point at the question of the common algorithm can＇ t suit for complex light condition such as backlight or front light, and real time and smooth are can ＇ t gave consideration at the same time, an auto exposure algorithm using image subregion and variable step size is given in the paper. The image is partitioned with the subject and background area, and the subject area is gave the more weight, and the step size is also adjusted according to the difference between the average brightness and the real brightness. The actual test results prove that the algorithm can get the perfect image in the complex light condition, as well as close to real brightness speedy and smoothly when the intensity of illumination changed.
出处
《电视技术》
北大核心
2015年第19期28-31,共4页
Video Engineering
关键词
自动曝光
图像多区域
变步长
主体区域
背景区域
auto exposure
image subregion
variable step
subject area
background area
