摘要
在JPEG2000图像中,采用灰度掩模方法对多个任意形状感兴趣区域进行标识时,由于其要求掩模的行数和列数和原图像相同,使得掩模的数据量非常庞大。为减小掩模的数据量,提出了一种块覆盖灰度掩模方法。该方法在研究灰度掩模的基础上,采用由多个像素点组成的分块灰度掩模来代替原单个像素点组成的灰度掩模,以对多个任意形状感兴趣区域进行标识。在对标识要求精度不太高的情况下,可极大地减小掩模的数据量。针对掩模扩张带给掩模数据量影响的问题,利用一个实例进行了数学计算,计算表明,当感兴趣区域面积足够大时,掩模扩张不会对掩模的数据量带来太大影响。
In the JPEG2000 image, when a number of arbitrary shaped Region of Interest （ROI） is marked with gray mask, the data quantity will get very large since the rows and columns must be the same as the o- riginal image. In order to reduce the data size, we put forward a new method of block covered gray mask. Based on the gray mask, when marking several arbitrary shaped ROIs, it adopted the block of gray mask that is made up of several pixels instead of single pixel. It will greatly reduce the data size when the demand of precision is not very high. An example was used to study the influence of extended mask on data quantity, which indicated that the mask extension cant cause too much effect on the data size when the area of ROI is big enough.
出处
《电光与控制》
北大核心
2009年第4期78-81,共4页
Electronics Optics & Control
