摘要
针对图像里处于复杂纹理背景中物体的轮廓提取正确率低的问题,首先研究了基于非经典感受野抑制的轮廓提取算法和HMAX模型,然后利用HMAX模型所具备的具有基本视皮层功能结构的优点,弥补了前者所依据的生物学视觉结构比较简单的不足,最后提出并实现了基于HMAX模型和非经典感受野抑制的轮廓提取算法。通过与Canny算子和非经典感受野抑制的轮廓提取算法的评估比较,表明本文算法有效提高了轮廓提取的正确率。
To solve the problem of low accuracy of contour detection of objects with complex texture background in the image,the existing contour detection algorithm based on non-classical receptive field inhibition and hierarchical model and X（HMAX） model was studied firstly.Then an improved contour detection algorithm based on HMAX model and non-classical receptive field inhibition was proposed and implemented.The HMAX model possesses the advantage of basic visual cortex functional structure.This compensates the oversimple shortcoming of biological visual structure,which the non-classical receptive inhibition contour detection algorithm is based on.The performance of the proposed algorithm is compared with Canny operators and non-classical receptive field inhibition contour detection algorithm.Results show that the improved algorithm can effectively increase the accuracy of contour detection.
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2012年第1期128-133,共6页
Journal of Jilin University:Engineering and Technology Edition
基金
国家自然科学基金项目(61101155)
吉林省科技发展计划项目(20101504)
吉林省教育厅'十一五'科学技术研究项目(2009604)
关键词
计算机应用
轮廓提取
非经典感受野
抑制
HMAX模型
computer applications
contour detection
non-classical receptive field
inhibition
hierarchical model and X（HMAX） model
