基于智能视觉的油画真伪鉴定技术研究  

Research on oil painting authenticity identification technology based on intelligent vision

作　　者：苏雪薇 SU Xuewei(Guangxi Normal University,Guilin 541001,China)

机构地区：[1]广西师范大学,广西桂林541001

出　　处：《现代电子技术》2020年第5期61-64,共4页Modern Electronics Technique

基　　金：2018年度广西高校中青年教师基础能力提升项目(2018KY0072)。

摘　　要：一种高精度、高效率的油画鉴定方法是鉴定作品真伪的辅助依据,可以提高油画鉴定的效率与可信性,因此提出基于智能视觉的油画真伪鉴定方法。构建智能视觉检测模型,获取油画图像特征;采用基于智能视觉的油画特征融合方法,融合油画特征的颜色和形状特征;计算油画特征差异系数与差异特征阈值,通过油画真伪鉴定规则实现油画真伪鉴定。研究结果验证,所提方法可以有效鉴别油画真伪,与专家鉴定方法和基于深度学习的真伪识别方法对比可知,该方法的鉴定准确率最高,鉴定时间开销最短,且抗干扰性最好,鉴定性能显著,具有较高的应用价值。An oil painting identification method with high precision and efficiency is an auxiliary basis for identifying the authenticity of oil painting works and can improve the efficiency and credibility of oil painting identification.In view of this,a method of oil painting authenticity identification based on intelligence vision is proposed.An intelligent vision detection model is constructed to obtain oil painting image features.The oil painting feature fusion method based on intelligent vision is adopted to fuse the color and shape features of oil paintings.The last step is to calculate the feature difference coefficient and the difference feature threshold value of oil paintings,and realize the authenticity identification by oil painting authenticity identification rules.The research results verify that the proposed method can effectively identify the authenticity of oil paintings.In comparison with the expert identification method and the authenticity identification method based on deep learning,the proposed method has the highest identification accuracy,the shortest identification time consumption and the best anti-interference performance.Therefore,the method has remarkable identification performance and high application value.

关 键 词：油画真伪鉴定 智能视觉 检测模型构建 油画特征获取 特征差异计算 油画特征融合 

分 类 号：TN911.1-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]