Firearm Classification using Neural Networks on Ring of Firing Pin Impression Images

This paper implements two layer neural networks with different feedforward backpropagation algorithms for better performance of firearm classification us-ing numerical features from the ring image. A total of 747 ring images which are extracted from centre of the firing pin impression have been captured from five different pistols of the Parabellum Vector SPI 9mm model. Then, based on finding from the previous studies, the six best geometric moments numerical fea-tures were extracted from those ring images. The elements of the dataset were further randomly divided into the training set (523 elements), testing set (112 el-ements) and validation set (112 elements) in accordance with the requirement of the supervised learning nature of the backpropagation neural network (BPNN). Empirical results show that a two layer BPNN with a 6-7-5 configura-tion and tansig/tansig transfer functions with ‘trainscg’ training algorithm has produced the best classification result of 98%. The classification result is an improvement compared to the previous studies as well as confirming that the ring image region contains useful information for firearm classification.

[1]  Bernard Widrow,et al.  Neural networks: applications in industry, business and science , 1994, CACM.

[2]  Zeno Geradts,et al.  Pattern recognition in a database of cartridge cases , 1999, Other Conferences.

[3]  Jun Kong,et al.  A Firearm Identification System Based on Neural Network , 2003, Australian Conference on Artificial Intelligence.

[4]  Dongguang Li A New Approach for Firearm Identification with Hierarchical Neural Networks Based on Cartridge Case Images , 2006, 2006 5th IEEE International Conference on Cognitive Informatics.

[5]  Jinsong Leng,et al.  On analysis of circle moments and texture features for cartridge images recognition , 2012, Expert Syst. Appl..

[6]  Cheng-Tan Tung,et al.  FISOFM: firearms identification based on SOFM model of neural network , 1994, 1994 Proceedings of IEEE International Carnahan Conference on Security Technology.

[7]  Clifton L. Smith,et al.  Optical imaging techniques for ballistics specimens to identify firearms , 1995, Proceedings The Institute of Electrical and Electronics Engineers. 29th Annual 1995 International Carnahan Conference on Security Technology.

[8]  Nor Azura Md Ghani,et al.  Analysis of Geometric Moments as Features for Identification of Forensic Ballistics Specimen , 2009, IWANN.

[9]  Timothy Masters,et al.  Practical neural network recipes in C , 1993 .

[10]  Gang Rong,et al.  A cartridge identification system for firearm authentication , 2000, WCC 2000 - ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000.

[11]  M S Bonfanti,et al.  The influence of manufacturing processes on the identification of bullets and cartridge cases--a review of the literature. , 1999, Science & justice : journal of the Forensic Science Society.

[12]  Mochammad Rif'an,et al.  The Implementation of Feedforward Backpropagation Algorithm for Digit Handwritten Recognition in a Xilinx Spartan-3 , 2012 .

[13]  Saadi Bin Ahmad Kamaruddin,et al.  Firearm recognition based on whole firing pin impression image via backpropagation neural network , 2011, 2011 International Conference on Pattern Analysis and Intelligence Robotics.

[14]  Dongguang Li,et al.  Intelligent Imaging of Forensic Ballistics Specimens for ID , 2008, 2008 Congress on Image and Signal Processing.

[15]  J Bijhold,et al.  Image matching algorithms for breech face marks and firing pins in a database of spent cartridge cases of firearms. , 2001, Forensic science international.