基于数据挖掘的学生档案照片像素细分算法  

Image segmentation algorithm for student file photos based on data mining

作　　者：邵泳兵 SHAO Yong-bing(Computer Department of Shantou Polytechnic,Shantou 515041,Guangdong Province,China)

机构地区：[1]汕头职业技术学院计算机系,广东汕头515041

出　　处：《信息技术》2020年第9期99-102,109,共5页Information Technology

摘　　要：为了提高学生档案照片分类识别和管理能力,提出基于数据挖掘的学生档案照片像素细分算法。设计学生档案照片采集样本数据库,结合学生档案照片像素点的特征分布进行边缘轮廓检测和特征区域像素重构,采用多维像素特征分解降噪方法抑制学生档案照片灰色像素特征点模糊集,提取学生档案照片序列几何特征点,进行学生档案照片的模块化区域划分处理,根据档案照片的人脸差异性进行学生档案照片的关键特征点定位,采用分块模板特征匹配方法进行学生档案照片区域重构,在仿射不变区域中进行学生档案照片像素细分和特征点匹配,采用图像分块模板匹配方法实现学生档案照片的自动像素细分识别。In order to improve the ability of classification,identification and management of student archives photos,a pixel subdivision algorithm of student archives photos based on data mining is proposed.The method comprises the following steps:designing a student file photo collection sample database,carrying out edge contour detection and feature region pixel reconstruction on the feature distribution of pixel points of the geometric student file photo,inhibiting fuzzy sets of gray pixel feature points of the student file photo by adopting a multi-dimensional pixel feature decomposition noise reduction method,extracting geometric feature points of the student file photo sequence,carrying out modular region division processing on the student file photo,and positioning key feature points of the student file photo according to face differences of the file photo.The method of block template feature matching is used to reconstruct the area of student file photos,pixel subdivision and feature point matching are carried out in the affine invariant area,and the method of image block template matching is used to realize automatic pixel subdivision identification of student file photos.

关 键 词：数据挖掘 学生档案 照片像素 细分模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]