摘要
针对图像视觉特征和情感语义之间的语义鸿沟,以图像纹理为低层特征,通过使用BP神经网络完成了图像低层特征到情感语义的映射;并在精度保持不变的前提下,对训练好的网络模型进行剪枝,最后通过神经网络规则抽取算法将隐含在神经网络模型中的知识转化为易于理解的IF-THEN规则形式。实验验证了方法的有效性和规则的可理解性。
To bridge the semantic gaps between the low-level image features and high-level emotional semantics,in the paper,it describes image features using texture and completes the semantic mapping through BP neural network.On the premise of keeping the accuracy of classification,the trained feedforward neural network is pruned using RX algorithm.Finally,the rules of IF-THEN which can be understood easily were extracted from pruned neural network model.The experiment shows the method in the paper is effective and the rules extracted are comprehensible.
出处
《系统仿真技术》
2010年第4期319-322,共4页
System Simulation Technology
关键词
图像纹理
情感语义
神经网络
规则抽取
image texture
affective semantic
neural network
rule extraction