Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis
We report an investigation into the application of the logistic regression classifier for estimation of sex in facial images. We used 2000 images, 1000 each of both sexes from a publicly available database and automatically detected facial landmarks and derived some morphometric facial indices. Thes...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-11-01
|
| Series: | Forensic Science International: Reports |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2665910721000578 |
| _version_ | 1819172939699847168 |
|---|---|
| author | Rajesh Verma Navdha Bhardwaj Pushap Deep Singh Arnav Bhavsar Vishal Sharma |
| author_facet | Rajesh Verma Navdha Bhardwaj Pushap Deep Singh Arnav Bhavsar Vishal Sharma |
| author_sort | Rajesh Verma |
| collection | DOAJ |
| description | We report an investigation into the application of the logistic regression classifier for estimation of sex in facial images. We used 2000 images, 1000 each of both sexes from a publicly available database and automatically detected facial landmarks and derived some morphometric facial indices. These indices were used as predictors for the classification. As the traditional manual extraction of facial landmarks is time consuming, automatic detection of the landmarks improves the efficiency. The logistic regression classification is also compared with two other classification methods, the likelihood-ratio (LR) based method where the features of a face are evaluated in terms of the probability distribution of these features in both the sexes, and the Convolutional Neural Networks (CNN) methods. While is former is desirable from the viewpoint of interpretability and to assess the strength of evidence, the latter is sophisticated. We report an AUC of 0.94 with true positive (TP) rate of 88.4% for males and 87.9% for females for logistic regression-based classification. This performance is better than the likelihood ratio classifier with TP rate of 79.6% for males and 82.2% for females. The overall performance of logistics regression is slightly less than the CNN classifier that has 89.3% TP rate for males and 92.6% for females. We have extended these models to a CCTV image database, more representative of the forensic scenario and found the logistic regression performing better than the CNN method on an average for 8 different types of cameras. We conclude that as a trade-off between simplicity and sophistication, the logistic regression classifier can be used for a two-class problem like classification of sex from facial morphometric indices, and that the likelihood ratio approach can assess the strength of the classification, in conformance with the requirements of evidence interpretation. |
| first_indexed | 2024-12-22T20:15:09Z |
| format | Article |
| id | doaj.art-763414fd42734c2693b3b8c07e8080bd |
| institution | Directory Open Access Journal |
| issn | 2665-9107 |
| language | English |
| last_indexed | 2024-12-22T20:15:09Z |
| publishDate | 2021-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Forensic Science International: Reports |
| spelling | doaj.art-763414fd42734c2693b3b8c07e8080bd2022-12-21T18:13:59ZengElsevierForensic Science International: Reports2665-91072021-11-014100226Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysisRajesh Verma0Navdha Bhardwaj1Pushap Deep Singh2Arnav Bhavsar3Vishal Sharma4Regional Forensic Science Laboratory, Mandi 175001, Himachal Pradesh, India; Corresponding authors.School of Computing and Electrical Engineering, Indian Institute of Technology, Kamand, Mandi 175005, Himachal Pradesh, IndiaSchool of Computing and Electrical Engineering, Indian Institute of Technology, Kamand, Mandi 175005, Himachal Pradesh, IndiaSchool of Computing and Electrical Engineering, Indian Institute of Technology, Kamand, Mandi 175005, Himachal Pradesh, IndiaInstitute of Forensic Science & Criminology, Panjab University, Chandigarh 160014, India; Corresponding authors.We report an investigation into the application of the logistic regression classifier for estimation of sex in facial images. We used 2000 images, 1000 each of both sexes from a publicly available database and automatically detected facial landmarks and derived some morphometric facial indices. These indices were used as predictors for the classification. As the traditional manual extraction of facial landmarks is time consuming, automatic detection of the landmarks improves the efficiency. The logistic regression classification is also compared with two other classification methods, the likelihood-ratio (LR) based method where the features of a face are evaluated in terms of the probability distribution of these features in both the sexes, and the Convolutional Neural Networks (CNN) methods. While is former is desirable from the viewpoint of interpretability and to assess the strength of evidence, the latter is sophisticated. We report an AUC of 0.94 with true positive (TP) rate of 88.4% for males and 87.9% for females for logistic regression-based classification. This performance is better than the likelihood ratio classifier with TP rate of 79.6% for males and 82.2% for females. The overall performance of logistics regression is slightly less than the CNN classifier that has 89.3% TP rate for males and 92.6% for females. We have extended these models to a CCTV image database, more representative of the forensic scenario and found the logistic regression performing better than the CNN method on an average for 8 different types of cameras. We conclude that as a trade-off between simplicity and sophistication, the logistic regression classifier can be used for a two-class problem like classification of sex from facial morphometric indices, and that the likelihood ratio approach can assess the strength of the classification, in conformance with the requirements of evidence interpretation.http://www.sciencedirect.com/science/article/pii/S2665910721000578Forensic scienceCCTV, sex estimationLikelihood ratioFacial landmarks, logistic regression |
| spellingShingle | Rajesh Verma Navdha Bhardwaj Pushap Deep Singh Arnav Bhavsar Vishal Sharma Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis Forensic Science International: Reports Forensic science CCTV, sex estimation Likelihood ratio Facial landmarks, logistic regression |
| title | Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis |
| title_full | Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis |
| title_fullStr | Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis |
| title_full_unstemmed | Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis |
| title_short | Estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis |
| title_sort | estimation of sex through morphometric landmark indices in facial images with strength of evidence in logistic regression analysis |
| topic | Forensic science CCTV, sex estimation Likelihood ratio Facial landmarks, logistic regression |
| url | http://www.sciencedirect.com/science/article/pii/S2665910721000578 |
| work_keys_str_mv | AT rajeshverma estimationofsexthroughmorphometriclandmarkindicesinfacialimageswithstrengthofevidenceinlogisticregressionanalysis AT navdhabhardwaj estimationofsexthroughmorphometriclandmarkindicesinfacialimageswithstrengthofevidenceinlogisticregressionanalysis AT pushapdeepsingh estimationofsexthroughmorphometriclandmarkindicesinfacialimageswithstrengthofevidenceinlogisticregressionanalysis AT arnavbhavsar estimationofsexthroughmorphometriclandmarkindicesinfacialimageswithstrengthofevidenceinlogisticregressionanalysis AT vishalsharma estimationofsexthroughmorphometriclandmarkindicesinfacialimageswithstrengthofevidenceinlogisticregressionanalysis |