Forensic Handwriting Identification Using Chicken Swarm Optimization-Enhanced Convolutional Neural Networks
Abstract
Forensic handwriting identification plays a pivotal role in criminal investigations and legal proceedings, requiring accurate and reliable techniques for analyzing handwriting samples. Convolutional Neural Networks (CNNs) are effective at feature extraction and classification for handwriting analysis, but their performance is highly sensitive to the choice of architectural hyperparameters, and manually tuned CNNs are prone to overfitting on forensic handwriting data. This study integrates the Chicken Swarm Optimization (CSO) algorithm with a CNN classifier (CSO-CNN) to automatically search the number of convolutional layers, the number and size of filters, and the batch size that minimize overfitting and maximize recognition accuracy. Handwritten samples were collected from staff and students of Ajayi Crowther University, Oyo, Nigeria, and organized into three variants (Original, Disguised, and Forged) with each variant split into training and test sets (72% and 28%, respectively) using a random-sampling hold-out procedure. CSO searched 30 candidate CNN configurations over the number of convolutional layers (1–3), the number of filters, the filter size, and the batch size, using recognition rate as the fitness function; the best-performing configuration used a single convolutional layer with 128 filters of size 3×3 and a batch size of 256. The resulting CSO-CNN model was compared against a standard, non-optimized CNN using accuracy, sensitivity, specificity, precision, false positive rate, and recognition time. CSO-CNN outperformed the standard CNN on every dataset variant: on the original dataset, accuracy improved from 82.14% to 86.43%; on the Disguised dataset, from 85.00% to 87.86%; and on the Forged dataset, from 83.57% to 89.29%, while recognition time was also reduced in all three cases. These results indicate that CSO-based hyperparameter search can meaningfully improve CNN-based forensic handwriting identification without requiring architectural redesign. As the evaluation reported here is based on a single hold-out split, repeated cross-validation and testing on independently collected datasets are recommended as a next step to confirm the statistical robustness and generalizability of these gains.References
1. Gavrilescu M., and Vizireanu N. (2018). Predicting the Big Five Personality Traits from Handwriting. Eurasip Journal on Image and Video Processing. 57, 1-17
2. Fenwa O. D., Omidiora E. O., Fakolujo O. A., Oke A. O. (2012). A modified Optical Backpropagation Neural Network Model in Online Handwritten Character Recognition System. International Journal of Computer Application Issue, 4(2); 190-201
3. Chaudhari K., Thakkar A., (2019) Survey on Handwriting- based Personality Trait Identification. Expert Systems with Applications 24, 282-308
4. Ojo O. S. (2024). Development of a Three Factor Authentication System for Online Banking. Ajayi Crowther Journal of Pure and Applied Sciences. Vol. 3 No. 2 pp1-11
5. Sudana O., Gunaya I. W., Putra I. G. (2020). Handwriting identification using Deep Convolutional Neural Network Method. TELKOMNIKA Telecommunication, Computing, Electronics and Control. 18 (4): 1934-1941
6. Caligiuri M. P., and Mohammed L. A. (2012). The Neuroscience of Handwriting: Applications for Forensic Document Examination. CRC Press. ISBN 966-3-340-94463-0
7. Kushki, A., Chau, T., and Anagnostou, E. (2011). Handwriting Difficulties in Children with Autism Spectrum Disorders: A scoping review. Journal of Autism and Developmental Disorders. 41 (12): 1706–1716
8. Rosenblum, S., Engel-Yeger, B. and Fogel, Y. (2013). Age-Related Changes in Executive Control And Their Relationships with Activity Performance in Handwriting. Human Movement Science. 32 (2): 363–376.
9. Szadvári, I., Ostatníková, D., & Durdiaková, J. B. (2023). Sex Differences Matter: Males and Females Are Equal but Not the Same. Physiology & Behavior, 259, 114038.
10. Ojo O. S., Oyediran M. O., Awodoye O. O., Ajagbe S. A., Awotunde J. B., Bandyopadhyaye A. and Adigun M. O. (2023). Real-time Face-based Gender Identification System Using Pelican Support Vector Machine. Elsevier/ Procedia Computer Science Vol. 42, pp: 197–211
11. Rafiepour, M., & Sartakhti, J. S. (2023). CTRAN: CNN-transformer-based network for natural language understanding. Engineering Applications of Artificial Intelligence, 126, 107013.
12. Kazheen I. T. and Adnan M. A. (2021). Deep Learning Convolutional Neural Network for Speech Recognition: A Review. International Journal of Science and Business. 5 (3): 1-14
13. Dulari B., Chirag P., Hardik T., Jigar P., Rasmika V., Sharnil P., Kirit M. and Hemant G. (2021) CNN Variants for Computer Vision: History, Architecture, Application, Challenges and Future Scope. Electronics (Switzerland), 10(20): 778-791
14. Oyediran M. O., Ojo O. S., Amole A. M., Jooda O. J. and Adedapo O. A. (2024). Optimizing Hyperparameters of a Deep Learning Algorithm for a Real Time Face Recognition System. Kongzhi yu Juece/Control and Decision. ISSN: 1001-0920 Volume 39, Issue 04, pp. 2653-2664
15. Ojo, O. S., Oyediran, M. O., Awodoye, O. O., Ajagbe, S. A., Awotunde, J. B., Bandyopadhyay, A., & Adigun, M. O. (2024). Real-time Face-based Gender Identification System Using Pelican Support Vector Machine. Procedia Computer Science, 235, 3236–3245. doi:https://doi.org/10.1016/j.procs.2024.04.306
16. Rastegar, H., & Giveki, D. (2023). Designing a new deep convolutional neural network for content-based image retrieval with relevance feedback. Computers and Electrical Engineering, 106, 108593.
17. Ying X. (2019). An Overview of Overfitting and its Solutions. Journal of Physics: Conference Series. 168(2): 1168-1192
18. Aguerchi, K., Jabrane, Y., Habba, M., & El Hassani, A. H. (2024). A CNN Hyperparameters Optimization Based on Particle Swarm Optimization for Mammography Breast Cancer Classification. Journal of Imaging, 10(2), 30.
19. Choudhary, A., Mian, T., Fatima, S., & Panigrahi, B. K. (2024). Nature‐inspired artificial bee colony‐based hyperparameter optimization of CNN for anomaly detection in induction motor. Expert Systems, 41(7), e13407.
20. Prathap, G., & Rathinasabapathy, R. (2023). A Text Summarization Hybrid Approach Using CNN and the Firefly Algorithm. SN Computer Science, 5(1), 119.
21. Deb S., Gao X., Tammi K., Kalita K. and Mahanta P. (2020). Recent Studies on Chicken Swarm Optimization algorithm: A Review (2014–2018). Artificial Intelligence Review, 53(3) 1737-1765,
22. Haji, L. M., Mustafa, O. M., Abdullah, S. A., & Ahmed, O. M. (2024). Enhanced Convolutional Neural Network for Fashion Classification. Engineering, Technology & Applied Science Research, 14(5), 16534–16538. https://doi.org/10.48084/etasr.8147
23. Giles, A. (2004). The Forensic Examination of Documents. White, Peter (ed.) Crime Scene to Court: The Essentials of Forensic Science. London Royal Society of Chemistry, 76, 142-171.
24. Fernandez-de-Sevilla R., Alonso-Fernandez F., Fierrez J. and Ortega-Garcia J. (2010); "Forensic Writer Identification using Allographic Features”. Proceedings of the 12th International Conference on Frontiers in Handwriting Recognition, 12, 308- 313
25. Okawa, M. (2023). Synergy of on-surface and in-air trajectories: Exploratory analysis of forensic online signatures implementing lessons learned from biometrics. Forensic Science International: Reports, 8, 100340.
26. Kundur, N. C., & Mallikarjuna, P. B. (2022). Deep Convolutional Neural Network Architecture for Plant Seedling Classification. Engineering, Technology & Applied Science Research, 12(6), 9464–9470. https://doi.org/10.48084/etasr.5282
27. Sudana, O., Bayupati, I. P. A., & Yudiana, D. G. (2020). Classification of Maturity Level of the Mangosteen using the Convolutional Neural Network (CNN) method. Int. J. Adv. Sci. Technol, 135(20).
28. Kumar, P., & Sharma, A. (2020). Segmentation-free writer identification based on Convolutional Neural Network. Computers & Electrical Engineering, 85, 106-116.
29. Ahlawat S., Choudhary A., Nayyar A., Singh S. and Yoon B. (2020); Improved Handwritten Digit Recognition Using Convolutional Neural Networks (CNN). Sensors Article, 20, 3335-3344. doi:10.3390/s20123344
30. Xiao M., Wu Y., Zuo G., Fan S., Yu H., Shaikh Z. A., and Wen Z. (2021). Addressing Overfitting Problem in Deep Learning-Based Solutions for Next Generation Data-Driven Networks. Wireless Communications and Mobile Computing, 10, 117-126
DOI:
https://doi.org/10.31449/inf.v50i15.15614Keywords:
Bio-Inspired Algorithm, Convolutional Neural Network (CNN), Chicken Swarm Optimization (CSO), Handwriting Identification, Pattern Recognition, ForensicForensic HandwritingDownloads
Published
Issue
Section
License
Authors retain copyright in their work. By submitting to and publishing with Informatica, authors grant the publisher (Slovene Society Informatika) the non-exclusive right to publish, reproduce, and distribute the article and to identify itself as the original publisher.
All articles are published under the Creative Commons Attribution license CC BY 3.0. Under this license, others may share and adapt the work for any purpose, provided appropriate credit is given and changes (if any) are indicated.
Authors may deposit and share the submitted version, accepted manuscript, and published version, provided the original publication in Informatica is properly cited.
