摘要
针对传统图像敏感信息识别方法精度不高的问题,提出一种基于机器学习的图像敏感信息识别方法。对识别图像进行去噪和细节锐化预处理操作,从颜色、形状和纹理三个角度提取预处理后的图像特征。结合所提取的图像进行图像敏感信息检索,通过图像特征提取图像敏感信息。根据机器学习中的支持向量机(SVM)原理,设计一种图像分类器实现对图像敏感信息的分类和识别。与传统的两种图像识别方法相比,该方法具有较高的识别精度,可在较短的识别时间完成图像敏感信息识别。
In view of the low recognition accuracy of the traditional image sensitive information recognition methods,a new method for image sensitive information recognition based on machine learning is proposed.The pre⁃processing operations for denoising and detail sharpening are performed on the image under recognition.The image features after pre⁃processing are extracted in the three aspects of color,shape and texture.The image sensitive information is retrieved in combination with the extracted image and extracted by the image features.On the basis of the principle of support vector machine(SVM)in machine learning,an image classifier is designed to realize the classification and recognition of image sensitive information.In comparison with the two traditional image recognition methods,this method has higher recognition precision and takes shorter recognition time for image sensitive information recognition.
作者
王海霞
李凯勇
WANG Haixia;LI Kaiyong(School of Physical and Electronic Information Engineering,Qinghai Nationalities University,Xining 810007,China)
出处
《现代电子技术》
2021年第19期66-70,共5页
Modern Electronics Technique
基金
青海省重点研发与转换计划(2019-GX-170)。
关键词
敏感信息识别
识别方法
图像信息检索
机器学习
信息提取
图像分类器
sensitive information recognition
recognition method
image information retrieval
machine learning
information extraction
image classifier
