Validation of forensic facial comparison by morphological analysis in photographic and CCTV samples

Global adoption of facial recognition technology with special reference to India-Present status and future recommendations

The face is the most essential part of the human body, and because of its distinctive traits, it is crucial for recognizing people. Facial recognition technology (FRT) is one of the most successful and fascinating technologies of the modern times. The world is moving towards contactless FRT after the COVID-19 pandemic. Due to its contactless biometric characteristics, FRT is becoming quite popular worldwide. Businesses are replacing conventional fingerprint scanners with artificial intelligence-based FRT, opening up enormous commercial prospects. Security and surveillance, authentication/access control systems, digital healthcare, photo retrieval, etc., are some sectors where its use has become essential. In the present communication, we presented the global adoption of FRT, its rising trend in the market, utilization of the technology in various sectors, its challenges and rising concerns with special reference to India and worldwide.

Prioritising quality: investigating the influence of image quality on forensic facial comparison

Morphological analysis in forensic facial comparison was recently validated for judicial use. However, no image quality assurance systems exist for this purpose, despite image triage being considered the best practice approach. Hence, this study aimed at testing a semi-quantitative scoring method to assess image quality and investigated facial image resolution and lighting quality quantitatively in a context of forensic facial comparison. For this purpose, 400 facial comparison photographic and CCTV image pools developed from the Wits Face Database were used. These facial images were analysed in prior studies that investigated the validity of morphological analysis. A semi-quantitative image quality scoring system was adapted and tested on the above sample and compared across correct and incorrect matches obtained as part of previous studies using a logistic regression model. In addition, facial images were cropped to the closest pixel comprising the face, head and neck areas; then, a face-to-image pixel proportion was calculated as an estimator of resolution quality; and pixel exposure qualities were obtained to be compared to facial comparison outcomes. Ideal and high image quality scores were related to correctness of matches, while low-quality scores were related to incorrect matches. High pixel proportions were related to true matches and low exposure was related to false positives, while high exposure was related to false negatives. These results suggest that an easy method for image triage could be employed by scoring image quality. Quantitative measures should be investigated further for thresholding quality suitability for confidence of facial comparisons.

The potential of facial nevi in personal identification

Forensic anthropologists dealing with personal identification (PI) of human remains have recently stressed the need to explore the potential of "secondary identifiers" for identifying victims who died in particular events for whom images often represent the main antemortem data available. Being the face the part most exposed in images, characteristics as pigmented skin lesions (PSLs), can be crucial if combined with other input. Since no data is available on frequencies and distribution of facial PSLs in the general population, this study aims at systematically collecting such data to verify their potential in PI and to open a debate on the aid that "secondary identifiers", regardless of their specific nature, can give to the identification of the deceased in specific forensic contexts. A retrospective analysis on three-dimensional facial models of 1039 Italian subjects (from 4 to 84 years old) was conducted to examine the incidence of PSLs discriminated according to size and position in well-defined facial areas. From the collected data we developed a probabilistic approach providing the likelihood ratio (LR) for two settings: (1) the relative frequencies of nevi in the various facial areas, providing the deriving compound probability of owning a certain facial PSLs pattern; and (2) codes describing the facial nevi pattern of each individual of our population, thus testing their uniqueness and so their potential in PI. The calculated LRs mostly proved high identifying strength, particularly when provided by the compound probability-based approach. Data on incidence and position of facial nevi, their generated codes, and the probabilistic approach here presented, all constitute a starting point for advancing secondary identifiers. Nonetheless, although this preliminary study proved facial PSLs as valuable and potentially useful for identification, their significance and validity should be interpreted with caution as we are still at the first theoretical step clearly based on ideal conditions, and thus further investigations are due on the limitations of their use in practical identifying settings. Therefore, being this systematic study only a preliminary one in its nature, it is recommended not to use this kind of approach until further studies will test its validity in several practical conditions.

High familiar faces have both eye recognition and holistic processing advantages

People recognize familiar faces better than unfamiliar faces. However, it remains unknown whether familiarity affects part-based and/or holistic processing. Wang et al., Frontiers in Psychology, 6, 559 (2015), Vision Research, 157, 89-96 (2019) found both enhanced part-based and holistic processing in eye relative to mouth regions (i.e., in a region-selective manner) for own-race and own-species faces, i.e., faces with more experience. Here, we examined the role of face familiarity in eyes (part-based, region-selective) and holistic processing. Face familiarity was tested at three levels: high-familiar (faces of students from the same department and the same class who attended almost all courses together), low-familiar (faces of students from the same department but different classes who attended some courses together), and unfamiliar (faces of schoolmates from different departments who seldom attended the same courses). Using the old/new task in Experiment 1, we found that participants recognized eyes of high-familiar faces better than low-familiar and unfamiliar ones, while similar performance was observed for mouths, indicating a region-selective, eyes familiarity effect. Using the "Perceptual field" paradigm in Experiment 2, we observed a stronger inversion effect for high-familiar faces, a weaker inversion effect for low-familiar faces, but a non-significant inversion effect for unfamiliar faces, indicating that face familiarity plays a role in holistic processing. Taken together, our results suggest that familiarity, like other experience-based variables (e.g., race and species), can improve both eye processing and holistic processing.

3D‐3D facial registration method applied to personal identification: Does it work with limited portions of faces? An experiment in ideal conditions

Personal identification of faces represents a challenging issue, especially for what concerns the quantification of the comparison. The 3D‐3D superimposition approach proved to distinguish between matches and mismatches. However, the potential of this procedure applied to cases where only parts of faces are visible still has to be verified. This study aimed at verifying the applicability of a 3D‐3D procedure to faces divided into three thirds. 3D models of fifty male subjects acquired through stereophotogrammetry were used. The 3D facial models were divided into upper, middle, and lower thirds and registered onto other models belonging to the same and different individuals according to the least point‐to‐point distance. In total, 50 matches and 50 mismatches were analyzed. RMS value (root mean square) of point‐to‐point distance between the two facial surfaces was calculated through VAM® software. Statistically significant differences between matches and mismatches in each facial third were assessed through Mann–Whitney test (p < 0.05). On average, RMS value in matches was 0.32 ± 0.12 mm in upper third, 0.36 ± 0.15 mm in middle third, and 0.40 ± 0.20 mm in lower third, respectively; in mismatches, RMS value was 1.40 ± 0.32 mm in upper third, 1.96 ± 0.58 mm in middle third, and 2.39 ± 0.90 mm in lower third, respectively. Differences in RMS values between matches and mismatches were significantly different for all facial thirds, without superimpositions (p < 0.01). This study shows that the existing 3D‐3D superimposition methods may be useful also when only a limited portion of face is visible in ideal conditions. Their application to forensic cases of identification still needs to be verified.

Forensic Facial Comparison: Current Status, Limitations, and Future Directions

Global escalation of crime has necessitated the use of digital imagery to aid the identification of perpetrators. Forensic facial comparison (FFC) is increasingly employed, often relying on poor-quality images. In the absence of standardized criteria, especially in terms of video recordings, verification of the methodology is needed. This paper addresses aspects of FFC, discussing relevant terminology, investigating the validity and reliability of the FISWG morphological feature list using a new South African database, and advising on standards for CCTV equipment. Suboptimal conditions, including poor resolution, unfavorable angle of incidence, color, and lighting, affected the accuracy of FFC. Morphological analysis of photographs, standard CCTV, and eye-level CCTV showed improved performance in a strict iteration analysis, but not when using analogue CCTV images. Therefore, both strict and lenient iterations should be conducted, but FFC must be abandoned when a strict iteration performs worse than a lenient one. This threshold ought to be applied to the specific CCTV equipment to determine its utility. Chance-corrected accuracy was the most representative measure of accuracy, as opposed to the commonly used hit rate. While the use of automated systems is increasing, trained human observer-based morphological analysis, using the FISWG feature list and an Analysis, Comparison, Evaluation, and Verification (ACE-V) approach, should be the primary method of facial comparison.

Performance of forensic facial comparison by morphological analysis across optimal and suboptimal CCTV settings

Facial comparison is an important yet understudied discipline in forensics. The recommended method for facial comparison in a forensic setting involves morphological analysis (MA) with the use of a facial feature list. The performance of this approach has not been tested across various closed-circuit television (CCTV) conditions. This is of particular concern as video and image data available to law enforcement is often varied and of subpar conditions. The present study aimed at testing MA across two types of CCTV data, representing ideal and less than ideal settings, also assessing which particular shortcomings arose from less-than-ideal settings. The study was conducted on a subset of the Wits Face Database arranged in a total of 225 face pools. Each face pool consisted of a target image obtained from either a high-definition digital CCTV camera or a low-definition analogue CCTV camera in monochrome, contrasted to 10 possible matches. The face pools were analysed and scored using MA and confusion matrices were used to analyse the outcomes. A notably high chance corrected accuracy (CCA) (97.3%) and reliability (0.969) was identified across the digital CCTV sample, while in the analogue CCTV sample MA appeared to underperform both in accuracy (CCA: 33.1%) and reliability (0.529). The majority of the errors in scoring resulted in false negatives in the analogue sample (75.2%), while across both CCTV conditions false positives were low (digital: 0.3%; analogue: 1.2%). Even though hit rates appeared deceptively high in the analogue sample, the various measures of performance used and particularly the chance corrected accuracy highlighted its shortfalls. Overall, CCTV recording quality appears closely associated to MA performance, despite the favourable error rates when using the Facial Identification Scientific Working Group feature list.

Development of the Wits Face Database: an African database of high-resolution facial photographs and multimodal closed-circuit television (CCTV) recordings

Forensic facial comparison is a commonly used, yet under-evaluated method employed in medicolegal contexts across the world. Testing the accuracy and reliability of facial comparisons requires large scale controlled and matching facial image databases. Databases that contain images of individuals on closed-circuit television (CCTV), with matching formal and informal photographs are needed for this type of research. Although many databases are available, the majority if not all are developed in order to improve facial recognition and face detection algorithms through machine learning, with very limited if any measure of standardisation. This paper aims to review the available databases and describe the development of a high resolution, standardised facial photograph and CCTV recording database of male Africans. The database is composed of a total of 6220 standardised and uncontrolled suboptimal facial photographs of 622 matching individuals in five different views, as well as corresponding CCTV footage of 334 individuals recorded under different realistic conditions. A detailed description of the composition and acquisition process of the database as well as its subdivisions and possible uses are provided. The challenges and limitations of developing this database are also highlighted, particularly with regard to obtaining CCTV video recordings and ethics for a database of faces. The application process to access the database is also briefly described.