• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.312-318
• /
• 2022

This paper proposes a model which classifies the type of guns and information about sound source location using deep neural network. The proposed classification model is composed of convolutional neural networks (CNN) and long short-term memory (LSTM). For training and test the model, we use the Gunshot Audio Forensic Dataset generated by the project supported by the National Institute of Justice (NIJ). The acoustic signals are transformed to Mel-Spectrogram and they are provided as learning and test data for the proposed model. The model is compared with the control model consisting of convolutional neural networks only. The proposed model shows high accuracy more than 90 %.

• The korean journal of orthodontics
• /
• v.53 no.3
• /
• pp.194-204
• /
• 2023

Objective: To investigate sex-specific correlations between the dimensions of permanent canines and the anterior Bolton ratio and to construct a statistical model capable of identifying the sex of an unknown subject. Methods: Odontometric data were collected from 121 plaster study models derived from Caucasian orthodontic patients aged 12-17 years at the pretreatment stage by measuring the dimensions of the permanent canines and Bolton's anterior ratio. Sixteen variables were collected for each subject: 12 dimensions of the permanent canines, sex, age, anterior Bolton ratio, and Angle's classification. Data were analyzed using inferential statistics, principal component analysis, and artificial neural network modeling. Results: Sex-specific differences were identified in all odontometric variables, and an artificial neural network model was prepared that used odontometric variables for predicting the sex of the participants with an accuracy of > 80%. This model can be applied for forensic purposes, and its accuracy can be further improved by adding data collected from new subjects or adding new variables for existing subjects. The improvement in the accuracy of the model was demonstrated by an increase in the percentage of accurate predictions from 72.0-78.1% to 77.8-85.7% after the anterior Bolton ratio and age were added. Conclusions: The described artificial neural network model combines forensic dentistry and orthodontics to improve subject recognition by expanding the initial space of odontometric variables and adding orthodontic parameters.

• Imaging Science in Dentistry
• /
• v.49 no.1
• /
• pp.19-26
• /
• 2019

Purpose: It has been proposed that using new prediction methods, such as neural networks based on dental data, could improve age estimation. This study aimed to assess the possibility of exploiting neural networks for estimating age by means of the pulp-to-tooth ratio in canines as a non-destructive, non-expensive, and accurate method. In addition, the predictive performance of neural networks was compared with that of a linear regression model. Materials and Methods: Three hundred subjects whose age ranged from 14 to 60 years and were well distributed among various age groups were included in the study. Two statistical software programs, SPSS 21 (IBM Corp., Armonk, NY, USA) and R, were used for statistical analyses. Results: The results indicated that the neural network model generally performed better than the regression model for estimation of age with pulp-to-tooth ratio data. The prediction errors of the developed neural network model were acceptable, with a root mean square error (RMSE) of 4.40 years and a mean absolute error (MAE) of 4.12 years for the unseen dataset. The prediction errors of the regression model were higher than those of the neural network, with an RMSE of 10.26 years and a MAE of 8.17 years for the test dataset. Conclusion: The neural network method showed relatively acceptable performance, with an MAE of 4.12 years. The application of neural networks creates new opportunities to obtain more accurate estimations of age in forensic research.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.2
• /
• pp.95-104
• /
• 2017

As the technology in smart device is growing and Social Network Services(SNS) are becoming more common, the data which is difficult to be processed by existing RDBMS are increasing. As a result of this, NoSQL databases are getting popular as an alternative for processing massive and unstructured data generated in real time. The demand for the technique of digital investigation of NoSQL databases is increasing as the businesses introducing NoSQL database in their system are increasing, although the technique of digital investigation of databases has been researched centered on RDMBS. New techniques of digital forensic investigation are needed as NoSQL Database has no schema to normalize and the storage method differs depending on the type of database and operation environment. Research on document-based database of NoSQL has been done but it is not applicable as itself to other types of NoSQL Database. Therefore, the way of operation and data model, grasp of operation environment, collection and analysis of artifacts and recovery technique of deleted data in HBase which is a NoSQL column-based database are presented in this paper. Also the proposed technique of digital forensic investigation to HBase is verified by an experimental scenario.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.2
• /
• pp.293-302
• /
• 2015

With the attack information collected during SMS phishing investigation, this paper will propose SMS phishing profiling model applying criminal profiling. Law enforcement agencies have used signature analysis by apk file hash and analysis of C&C IP address inserted in the malware. However, recently law enforcement agencies are facing the challenges such as signature diversification or code obfuscation. In order to overcome these problems, this paper examined 169 criminal cases and found out that 89% of serial number in cert.rsa and 80% of permission file was reused in different cases. Therefore, the proposed SMS phishing profiling model is mainly based on signature serial number and permission file hash. In addition, this model complements the conventional file hash clustering method and uses code similarity verification to ensure reliability.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.3
• /
• pp.417-428
• /
• 2020

With the rapid development of information and communication technology, mobile devices have become an essential tool in our lives. Mobile devices are used as important evidence in criminal proof, as they accumulate data simultaneously with PIM functions while working with users most of the time. The mobile forensics is a procedure for obtaining digital evidence from mobile devices and should be collected and analyzed in accordance with due process, just like other evidence, and the integrity of the evidence is essential because it has aspects that are easy to manipulate and delete. Also, the adoption of evidence relies on the judges' liberalism, which necessitates the presentation of generalized procedures. In this paper, a mobile forensics model is presented to ensure integrity through the generalization of procedures. It is expected that the proposed mobile forensics model will contribute to the formation of judges by ensuring the reliability and authenticity of evidence.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.647-659
• /
• 2023

As digital evidence becomes more important in criminal investigations, disputes are increasing in court. As media diversifies and the scope of analysis expands, the level of expertise in digital forensics is also increasing. However, no competency model has been developed to define the capabilities of digital evidence examiners or to judge their expertise. There have been some studies that have derived the capabilities necessary for digital evidence examiner, but they are still insufficient. Therefore, in this study, 25 competency evaluation factors in a total of 9 competency groups were defined using methodologies such as expert FGI and Delphi survey. Specifically, it was defined as Digital Forensics Theory, Digital Evidence Collection&Management, Disk Forensics, Mobile Forensics, Video Forensics, infringement forensics, DB Forensics, Embedded(IoT) Forensics, and Cloud Forensics. The digital evidence examiner competency model is expected to be used in various fields such as recruitment, education and training, and performance evaluation in the future.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.6
• /
• pp.55-65
• /
• 2011

Recently, use of cloud computing service is dramatically increasing due to wired and wireless communications network diffusion in a field of high performance Internet technique. Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. In a view of digital forensic investigation, it is difficult to obtain data from cloud computing service environments. therefore, this paper suggests analysis method of AWS(Amazon Web Service) and Rackspace which take most part in cloud computing service where IaaS formats presented for data acquisition in order to get an evidence.

• International Journal of Computer Science & Network Security
• /
• v.24 no.6
• /
• pp.207-215
• /
• 2024

Cloud computing is now used by most companies, business centres and academic institutions to embrace new computer technology. Cloud Service Providers (CSPs) are limited to certain services, missing some of the assets requested by their customers, it means that different clouds need to interconnect to share resources and interoperate between them. The clouds may be interconnected in different characteristics and systems, and the network may be vulnerable to volatility or interference. While information technology and cloud computing are also advancing to accommodate the growing worldwide application, criminals use cyberspace to perform cybercrimes. Cloud services deployment is becoming highly prone to threats and intrusions. The unauthorised access or destruction of records yields significant catastrophic losses to organisations or agencies. Human intervention and Physical devices are not enough for protection and monitoring of cloud services; therefore, there is a need for more efficient design for cyber defence that is adaptable, flexible, robust and able to detect dangerous cybercrime such as a Denial of Service (DOS) and Distributed Denial of Service (DDOS) in heterogeneous cloud computing platforms and make essential real-time decisions for forensic investigation. This paper aims to develop a framework for digital forensic for the detection of cybercrime in a joined heterogeneous cloud setup. We developed a Digital Forensics model in this paper that can function in heterogeneous joint clouds. We used Unified Modeling Language (UML) specifically activity diagram in designing the proposed framework, then for deployment, we used an architectural modelling system in developing a framework. We developed an activity diagram that can accommodate the variability and complexities of the clouds when handling inter-cloud resources.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.25 no.3
• /
• pp.374-379
• /
• 2017

In vehicle to pedestrian accidents, cracks occur in the vehicle laminated glass due to impact of a pedestrian's head. In this study, FMH(Free Motion Headform) was used to experiment on and analyze the crack patterns on a vehicle laminated glass that collides with an adult headform at speeds of 20 km/h, 30 km/h, and 40 km/h, respectively. Applying the acquired experimental data and material property of the vehicle laminated glass to the structural analysis program LS-Dyna, we could develop the FE model of vehicle laminated glass similar to real vehicle laminated glass. We could estimate the head impact velocity and pedestrian's vehicle impact velocity using the Madymo program.