•  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
期刊检索


检索
检索项:
检索词:
总目录
  • WCSB9︱2019(第九届)世界采样和混样大会 查看全文>>
  • 关于召开第十五届全国青年分析测试学术报告会的通知 查看全文>>
友情链接
基于滤波器-光谱数据降维的指甲地区识别
    点此下载全文
作者单位
古锟山,王继芬,曾啸虎 1. 中国人民公安大学 侦查学院北京 1000382. 酒泉卫星发射中心甘肃 酒泉 735000 
基金项目:中央高校基本科研业务费专项资金资助(2021JKF208)
中文摘要:该文从实际案件中收集了5个地区共计204份指甲样本,运用希尔伯特变换滤波器对原始谱图进行降噪处理,然后采用主成分分析进行数据降维,借助朴素贝叶斯、随机森林以及偏最小二乘判别分析模型开展指甲地区的识别工作,并根据模型的识别率和相关指标筛选出最佳预处理方法和最优识别模型。结果表明,经预处理后的原始谱图识别率得到显著提升,希尔伯特变换滤波器结合主成分分析是最佳预处理方法,随机森林模型的稳定性和识别率均高于朴素贝叶斯和偏最小二乘判别分析模型,对最佳预处理方法的训练集识别率为94.88%,测试集识别率为93.47%。该方法能有效降低谱图的噪声,减少数据的冗余,提高模型的识别效果,为法庭科学中指甲地区的快速鉴定提供了参考。
中文关键词:光谱学  ?指甲  ?希尔伯特变换滤波器  ?主成分分析  ?机器学习
 
Recognition of Fingernail Region Based on Filter-Spectral Feature Extraction
Abstract:Fingernail is one of the common biological evidences on the scene of the case.The rapid inspection of fingernails found in the scene could provide direction and clues for case investigation.Meanwhile,application of machine learning for quick and nondestructive detection of the testing material is an important branch of court science.Filter could effectively remove the noise and background interference of the spectra.The dimension reduction of the spectral data could effectively reduce the dimension of the data,and improve the recognition effect of the model.In this paper,a total of 204 nail samples from the actual cases of five regions were collected.The original spectra were denoised by Hilbert transform filter(HTF),and then the principal component analysis(PCA)was used to reduce the dimension of the original data and the denoised data.Naive Bayes(NB),random forest(RF) and partial least squares discriminant analysis(PLS-DA) model were used to carry out the identification of nail area.According to the recognition rate and related indicators of the model,the optimal preprocessing method and optimal recognition model for nail area identification were selected.The results demonstrated that the recognition rate of the original spectra is significantly improved after preprocessing.HTF combined with PCA is the best preprocessing method.The recognition rate of RF for the training set of the best pretreatment method is 94.88%,while that for the test set is 93.47%.This method could effectively reduce the noise of spectra,reduce the redundancy of data,improve the recognition effect of the model,and provide some reference for the rapid identification of nail areas in forensic science.
Key Words:spectroscopy  ?fingernail  ?Hilbert transform filter  ?principal component analysis(PCA)  ?machine learning
引用本文:古锟山,王继芬,曾啸虎.基于滤波器-光谱数据降维的指甲地区识别[J].分析测试学报,2022,41(5):746-753.
摘要点击次数: 783
全文下载次数: 561
查看全文  下载PDF阅读器