System Recognizes the Digital Image of Pistol Shell Casings by Developing Algorithms Combined with Deep Learning

Aree Jivorarak, K. Meethongjan, Narong Kunides
{"title":"System Recognizes the Digital Image of Pistol Shell Casings by Developing Algorithms Combined with Deep Learning","authors":"Aree Jivorarak, K. Meethongjan, Narong Kunides","doi":"10.53848/ssstj.v10i2.515","DOIUrl":null,"url":null,"abstract":"Gun-related violence in Thailand is in a high rate. Resume reports showed that most of them caused by gun-shooting.Thus, Firearms and bullets are important evidence in the judicial process to link the events and the perpetrators.Therefore, the aim of this study was to present the system recognizes the digital image of pistol shell casings by developing algorithms combined with deep learning. The objectives of this forensic study were to 1) analyze, design, and develop a Pistol Identification System (PIS) based on breech face marks of cartridge case digital images, and 2) achieve a guideline or an alternative method for facilitating an expert to investigate firearms linked to the offender. In this research the PIS that was designed with programming language applied to develop algorithms for identification of the breech face marks of cartridge case digital images. In addition to that, MATLAB’s tools were applied in the deep learning process to achieve the final PIS model. The steps of deep learning technique were composed of designing a training and repeat the experiments over multiple cycles (Epoch) for the purpose of confirming, test and adjust the proportions of the hidden layers until reaching the ratio of 80:10:10 and accomplishing a satisfied averaged accuracy rate. The PIS model was subsequently used for comparison and predict the image pair through database management technology. Materials used in this study were composed of 50,000 images of rear plates of .38 Cartridge case, Camera, Mobile Phone, Computer, MATLAB language and Microsoft Access software. The findings showed that the PIS developed is of satisfactory accuracy capable of accurately matching the pairs of images stored in the database and could also be traced back to the gun used at the scene and gun owners. The results of this study would apply as the alternative or guideline to PIS and even would help forensic practitioners to cross-checking and investigating firearms in relation to the offender.","PeriodicalId":31349,"journal":{"name":"Suan Sunandha Rajabhat University Journal of Science and Technology","volume":"30 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Suan Sunandha Rajabhat University Journal of Science and Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.53848/ssstj.v10i2.515","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Gun-related violence in Thailand is in a high rate. Resume reports showed that most of them caused by gun-shooting.Thus, Firearms and bullets are important evidence in the judicial process to link the events and the perpetrators.Therefore, the aim of this study was to present the system recognizes the digital image of pistol shell casings by developing algorithms combined with deep learning. The objectives of this forensic study were to 1) analyze, design, and develop a Pistol Identification System (PIS) based on breech face marks of cartridge case digital images, and 2) achieve a guideline or an alternative method for facilitating an expert to investigate firearms linked to the offender. In this research the PIS that was designed with programming language applied to develop algorithms for identification of the breech face marks of cartridge case digital images. In addition to that, MATLAB’s tools were applied in the deep learning process to achieve the final PIS model. The steps of deep learning technique were composed of designing a training and repeat the experiments over multiple cycles (Epoch) for the purpose of confirming, test and adjust the proportions of the hidden layers until reaching the ratio of 80:10:10 and accomplishing a satisfied averaged accuracy rate. The PIS model was subsequently used for comparison and predict the image pair through database management technology. Materials used in this study were composed of 50,000 images of rear plates of .38 Cartridge case, Camera, Mobile Phone, Computer, MATLAB language and Microsoft Access software. The findings showed that the PIS developed is of satisfactory accuracy capable of accurately matching the pairs of images stored in the database and could also be traced back to the gun used at the scene and gun owners. The results of this study would apply as the alternative or guideline to PIS and even would help forensic practitioners to cross-checking and investigating firearms in relation to the offender.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
该系统通过开发与深度学习相结合的算法对手枪弹壳的数字图像进行识别
泰国与枪支有关的暴力事件发生率很高。简历报告显示,大多数是由枪击造成的。因此,枪支和子弹是司法程序中将事件和肇事者联系起来的重要证据。因此,本研究的目的是通过开发与深度学习相结合的算法,提出系统识别手枪弹壳的数字图像。这项法医研究的目标是:1)分析、设计和开发基于弹壳数字图像的后膛标记的手枪识别系统(PIS); 2)实现指导方针或替代方法,以方便专家调查与罪犯有关的枪支。本研究利用编程语言设计的PIS,开发了弹壳数字图像后膛痕迹识别算法。除此之外,在深度学习过程中应用MATLAB的工具来实现最终的PIS模型。深度学习技术的步骤是设计一个训练,在多个循环(Epoch)上重复实验,以确认、测试和调整隐藏层的比例,直到达到80:10:10的比例,并获得满意的平均准确率。随后利用PIS模型进行比对,并通过数据库管理技术对图像对进行预测。本研究使用的材料由5万张。38弹壳后板图像、相机、手机、计算机、MATLAB语言和Microsoft Access软件组成。结果表明,PIS系统具有令人满意的精度,能够准确匹配数据库中存储的成对图像,并且还可以追溯到现场使用的枪支和枪支所有者。这项研究的结果可以作为PIS的替代或指导,甚至可以帮助法医从业者交叉检查和调查与罪犯有关的枪支。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
审稿时长
8 weeks
期刊最新文献
Analysis and Design of Building’s Structural Members (Slabs and Beams) using Microsoft Excel and AUTOCAD Software: A Case Study of MARCGSO’ Building Assessment of Fermented Rice Straw as Nutritive Substances for Bio-based Plant Pot System Recognizes the Digital Image of Pistol Shell Casings by Developing Algorithms Combined with Deep Learning XANES, XPS and Raman Studies of Hafnium Oxide Thin Films fabricated by RF Magnetron Sputtering at Different Power Influence of annealing and etching on physical and wetting properties of acrylic surface
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1