Abstract
为了适应图像分割编码的需要,提高编码性能和效率,本文研究了一种图像区域分割新方法.源于人眼成像原理和视神经网络的知觉分割特性,首先提出一种具有脉冲耦合和梯度锐化能力的神经元网络模型.然后通过构造一个拟合函数对相邻神经元的相似刺激输入进行平滑处理,而对具有不连续变化特性的刺激输入进行锐化,使得神经元比较容易地感知到均匀亮度区域和目标边缘的准确位置.最后通过实验验证了该算法的有效性.本文算法能够准确、有效地的分割出均匀区域,并且与原始图像具有很好的对应关系.在将本文算法应用到图像区域分割编码中,能够大大提高编码的效率,并得到高质量的重建图像.
Keywords:
Process (computing) Identification (biology) Product (mathematics)
Metrics
1
Cited By
0.48
FWCI (Field Weighted Citation Impact)
0
Refs
0.80
Citation Normalized Percentile
Is in top 1%
Is in top 10%
Citation History
Topics
Artificial Intelligence in Games
Physical Sciences → Computer Science → Artificial Intelligence
Language and cultural evolution
Social Sciences → Social Sciences → Cultural Studies
Computability, Logic, AI Algorithms
Physical Sciences → Computer Science → Computational Theory and Mathematics
Related Documents
JOURNAL ARTICLE
基于分割的彩色图像编码
Journal: Science in China Series F-Information Sciences (in Chinese) Year: 2009 Vol: 39 (4)Pages: 405-415
JOURNAL ARTICLE
一种新的高分辨率图像的聚类分割方法
Journal: Chinese Science Bulletin (Chinese Version) Year: 2012 Vol: 57 (34)Pages: 3334-3334
JOURNAL ARTICLE
一种基于清晰区域筛选的图像增强方法
张谦谦 Zhang Qianqian马卫红 Ma Weihong
Journal: Laser & Optoelectronics Progress Year: 2024 Vol: 61 (22)Pages: 2237009-2237009
JOURNAL ARTICLE
一种基于编码单元快速划分的VVC帧内编码方法
Journal: DOAJ (DOAJ: Directory of Open Access Journals) Year: 2024