Hands and Palms Recognition by Transfer Learning for Forensics: A Comparative Study
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
In the realm of forensic science, precise identification of individuals holds paramount importance in both investigative procedures and legal proceedings. Hands and palms recognition has emerged as a valuable biometric modality within forensic applications, owing to the distinct and intricate features inherent to these anatomical regions. The elaborate patterns of veins, creases, and ridges present on palms and fingers serve as rich sources of biometric data, crucial for accurate identification purposes. Furthermore, given the frequent involvement of hands and palms in criminal activities such as theft and assault, their recognition becomes imperative for establishing links between suspects and crime scenes. However, developing robust recognition systems tailored for forensic applications poses notable challenges, including variations in hand poses, lighting conditions, and image quality. To address these hurdles, sophisticated deep learning techniques, notably transfer learning, have been employed. By harnessing pre-trained deep learning models namely NasNetLarge, NasNetMobile, and EfficientNet, initially trained on expansive datasets for general image recognition tasks, we can adapt these models to the specific task of hands and palms recognition in forensic contexts. Our findings reveal that all three models consistently achieved over 92% accuracy across all metrics evaluated, demonstrating their efficacy as strong contenders for the hands-and-palms recognition task. Notably, the EfficientNet model exhibited superior performance compared to its counterparts, boasting more than 95.8% accuracy, precision, F1-score and recall, along with more than 98.6% specificity and 99.4% AUC.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it