摘要
一直以来,视觉图像都是人们从外界获取的一种重要信息来源。图像分割是针对图像像素的颜色、空间和纹理等底层语义信息对图像进行分割。超像素图像分割以超像素块代替像素作为图像处理的基本单元,减少了后续算法的计算量。但超像素算法无法很好地兼顾实时处理和处理精度,因而提升处理效率和精度将是超像素图像分割算法需要持续研究和改进的重要任务。
Visual images have always been an important source of information for people to obtain from the outside world.Image segmentation is to segment an image based on the underlying semantic information such as color,space,and texture of image pixels.Superpixel image segmentation uses superpixel blocks instead of pixels as the basic unit of image processing,which reduces the computational complexity of subsequent algorithms.Superpixel algorithm cannot take into account real-time processing and processing accuracy well,and improving processing efficiency and accuracy will be an important task for superpixel image segmentation algorithms that need continuous research and improvement.
作者
方堃
谢淑丽
齐微微
王伯燕
王锐
姚青
FANG Kun;XIE Shuli;QI Weiwei;WANG Boyan;WANG Rui;YAO Qing(Ningbo FOTILE Kitchenware Co.,Ltd.,Ningbo 315336;Key Laboratory of Healthy&Intelligent Kitchen System Integration of Zhejiang Province,Ningbo 315336;CHEARI(Beijing)Certification&Testing Co.,Ltd.,Beijing 100176)
出处
《家电科技》
2022年第S01期604-607,共4页
Journal of Appliance Science & Technology
关键词
超像素
图像分割
遥感影像
智能家电
Superpixel
Image segmentation
Remote sensing image
Smart appliances