Earticle

This study proposes and implements a blockchain based system to ensure the integrity of digital evidence. Digital evidence plays a crucial role in legal disputes and criminal investigations, However conventional centralized systems are vulnerable to data tampering and loss. To address these challenges, this research designs a system that guarantees the integrity of digital evidence throughout the processes of collection, storage, and verification by utilizing a private blockchain. This paper begin by examining blockchain and various consensus algorithms, followed by a detailed description of the implementation methods and processes of the proposed system. The system employs Ethereum-based GoQuorum to construct the private blockchain. Also, the QBFT consensus algorithm used in this network provides high security and fast processing speeds. By developing the smart contract, we enabled the storage and retrieval of the generated hash value of digital evidence on the blockchain. Our System verifies integrity by comparing the hash value stored on the chain with the hash value of the original image. Consequently, the blockchain-based integrity verification system has contributed to enhancing the reliablility of digital evidence and has improved both efficiency and security in legal proceedings.

The quality of latent fingerprint images can vary significantly based on deposition conditions, enhancement techniques, and imaging environments. This study investigates deep learning-based enhancement techniques for improving low-quality latent fingerprint images. GAN-based methods were excluded due to concerns regarding potential ridge distortion and artifact generation, leading to the selection of the Very Deep Super-Resolution (VDSR) model, which is based on a CNN architecture specialized for image processing. Additionally, VDSR offers the advantage of processing fluorescent images without the need for preprocessing. The VDSR model was modified to enable direct input of low-quality images, and three versions were evaluated: the pre-trained model, a fine-tuned model trained with an additional 100 ink fingerprint samples, and a newly-trained model using a dataset of 5,000 ink fingerprints. The pre-trained model, which yielded the best results, was compared with other enhancement methods, including R-ESRGAN 4x, a GAN-based image upscaler from Stable Diffusion, and traditional techniques such as intensity index adjustment and spatial filtering implemented through ImagePro. In general, deep learning-based techniques outperformed traditional methods, with VDSR proving particularly effective for enhancing fingerprint images. A combined approach that integrates deep learning-based techniques with traditional methods is expected to hold promising potential for further improving the quality of latent fingerprint images.

Biological evidence collected at crime scenes plays a crucial role in confirming the identities of individuals. Blood, particularly is a commonly encountered bodily fluid at such scenes, and fluids containing blood, such as saliva, semen, vaginal fluid, and urine, are excellent tools for personal identification. The accurate identification and collection of blood is essential for reconstructing crime scenes and subsequently identifying suspects. Blood is commonly found at crime scenes distributed across various surfaces and objects, serving as important evidence. The analysis of bloodstains provides valuable information about the events that occurred during a crime, and bloodstain analysis is a vital tool for reconstructing events at a crime scene. Initially, the presence of blood is confirmed based on factors such as time, conditions, and environment. Invisible bloodstains are identified by visually inspecting suspicious areas, followed by performing serological preliminary and confirmatory tests to determine the presence of blood. Notably, luminol and KM reagents are commonly used reagents at crime scenes. In cases where bloodstain analysis indicates a positive result, personal identification is conducted using nucleic acid (DNA) alongside confirmatory tests. Experimental results show that the luminol reagent exhibits higher sensitivity than the KM reagent. During DNA amplification, it is observed that factors such as time, conditions, and environment influence DNA amplification more than reagents. Furthermore, the STR marker D1S80 showed a more significant reduction compared to the TH01 marker. This experiment holds potential applicability in various crime scenes, offering foundational data regarding results based on sample types, temperature, environment, and time.

Even though vehicles are important means of transportation in modern society, they are accompanied by lots of traffic accidents. Sudden unintended acceleration(SUA) is supposed to be one of them. This study investigates the effect of lubricating system of vehicle equipped with turbocharger on SUA from two perspectives which are the quality and quantity of the lubricating oil system. The diluted lubricating oil with diesel oil didn't show deteriorating effect on the consistency of turbocharger at normal quantity. However, excessive quantity of oil showed dramatic effect on engine performance of exhausting white smoke with noise and vibration. This malfunction was thought to be from the flow of lubricating oil through blowby gas ventilation system into combustion chamber. Lest the problem should happen, lubrication management of vehicles is thought to be necessary for both sides of the quality and quantity of the lubricating oil.

When washing fabrics with blood on them, it is important to know the factors that affect the washing characteristics of blood. To investigate this, the authors prepared different thickness fabrics of 100% cotton, cotton/polyester blend(65%/35%), cotton/polyester blend(35%/65%), and 100% polyester. The cloths were then contaminated with blood, washed with detergent, and visualized with acidic hydrogen peroxide(a blood sensitive reagent) to observe the amount of blood retained. The results showed that the blood washed out better with more washing cycles, and the blood washed out better with a lower percentage of cotton. However, regardless of the blend, the thickness of the fabric did not seem to have any correlation with the degree of blood washing. On the other hand, if the fabric surface was coated, the blood washed well.

STR typing of crime scene evidence is currently contributing to solving various cases. In a autopsy case, the identity of the body is confirmed by collecting blood and comparing the STR typing results of the deceased with those of the family. This study, the rapid DNA extraction reagent was used to purify DNA from autopsy samples, blood and costal cartilage, and the effect of the rapid DNA extraction reagent on STR type was examined by comparing it with standard DNA. As a result of the experiment, the average DNA concentration using the rapid extraction reagent was purified to 0.277 ng/㎕ for blood and 0.069 ng/㎕ for costal cartilage, which met the recommended DNA concentration of 0.050 ng/㎕ for the STR amplification kit. The DNA degradation index was confirmed to be a maximum of 1.703 for blood and 0.660 for costal cartilage. In the case of STR type detection, DNA types were detected in all blood and costal cartilage samples. In the balance comparison of the values of heterozygous loci, there was no significant difference in the analysis values compared to the standard DNA samples, with 3 loci in blood, 6 loci in costal cartilage, and 6 loci in 007.

This study analyzes parking accident types and evaluates the performance of the Parking Autonomous Emergency Braking (P-AEB) system. A total of 4,071 real-world black box accident cases were analyzed, revealing that rearward parking accidents occur approximately four times more frequently than forward ones. Additionally, 79.1% of parking accidents involved collisions with stationary vehicles. The most common accident types were non-steering rearward collisions (34.0%), steering rearward diagonal collisions (24.2%), and non-steering diagonal rearward collisions (16.2%). The P-AEB performance evaluation followed the RCAR guidelines, testing 18 scenarios (6 forward, 12 rearward collisions). Results indicate that higher speeds increase the likelihood of rearward collisions, and all vehicles failed to avoid collisions in diagonal rearward parking scenarios. Additionally, P-AEB did not activate when entering forward while turning left, resulting in collisions. This study confirms that P-AEB performance varies depending on the installed system and parking environment, and in some situations, current technology may not fully prevent accidents. Future research should focus on enhancing P-AEB sensor detection and evaluation methods to improve real-world accident prevention effectiveness.

According to statistics from the National Forensic Service (NFS), the number of telephone financial fraud cases has more than doubled since onset of COVID-19 compared to the previous year. Telephone financial fraud, including false emergency calls to numbers 112 or 119, involves using telecommunication networks to threaten victims or extort money and valuables. The amount lost to voice phishing, in particular, has been steadily increasing annually posing a significant social problem. In such cases of telephone financial fraud, voice recordings serve as a crucial piece of evidence. This paper researches the forensic application of the latest deep learning-based speaker recognition technology for analyzing an individual from voice evidence. It explains the operational principles and usage of VOCALISE, a commercial speaker recognition engine developed by Oxford Wave Research. Notably, VOCALISE, widely used in forensic analysis organizations worldwide, is adept at creating sample groups from speaker datasets, measuring their statistical characteristics, and analyzing specific voice signals. To evaluate performance on korean speech data, we used 736 voice files collected from 159 individuals involved in actual voice crime cases to calculate the Equal Error Rate (EER) and Cost of Log-Likelihood Ratio (CLLR). Based on these real audio recordings when actual voice evidence is submitted by investigative agencies the paper explains how to use VOCALISE for a globally accepted likelihood ratio-based verbal expression method.

DNA methylation is a promising biomarker for forensic age estimation due to its strong correlation with chronological age. This study developed a droplet digital PCR (ddPCR) assay targeting three age-associated markers— ELOVL2, FHL2, and TRIM59—to evaluate its application in forensic cases. The assays were optimized using 100% methylated and unmethylated control DNA to evaluate probe specificity and sensitivity to methylation percentage. The assays exhibited high specificity, successfully distinguishing between methylated and unmethylated targets, and maintained sensitivity even at low methylation levels. Moreover, blood samples from three Korean individuals (aged 25, 42, and 55 years) were analyzed, revealing a clear age-related increase in methylation levels across all markers, with ELOVL2 exhibiting the strongest relation, followed by FHL2 and TRIM59. Reproducibility was evaluated by duplicate testing, with coefficient of variation values below 15% for all markers. These findings highlight the potential application of this ddPCR assay for forensic age estimation.

This study utilized scanning electron microscopy (SEM) to analyze the morphological characteristics of hair cut by various methods. In forensic investigations, hair serves as crucial evidence. When the hair root is present, DNA analysis can be employed for individual identification. However, when hair is cut in the middle, DNA analysis becomes challenging, making the morphological analysis of cut marks an essential aspect of forensic examinations. In this study, hair samples were cut using various tools, including a scalpel, utility knife, and scissors. SEM images were then obtained, and the morphological differences between tool-cut hair and hair severed by tensile force were compared and analyzed. The results indicated that tool-cut hair exhibited distinct striation marks on the cross-section, with the cuticle layer shifting in the direction of tool insertion, leaving scratch-like marks. In contrast, hair that was forcibly pulled apart showed no tool marks and displayed an irregular fracture pattern. These findings contribute to the quantitative and qualitative assessment of hair cut marks in forensic investigations and provide fundamental data for expanding the application of morphological hair analysis in criminal cases.

This study aims to determine the optimal gold evaporation time in Vacuum Metal Deposition (VMD) processes used for developing latent fingerprints on non-porous surfaces. VMD is a highly sensitive technique that visualizes latent fingerprints through the sequential deposition of gold and zinc under high vacuum conditions. In this research, the effects of gold evaporation duration were analyzed on two types of surfaces: glass slides and PVC book covers. Using the IDENTICOAT 500 VMD system, gold wires and zinc plates were vaporized under controlled conditions, and fingerprint development was evaluated at three different gold evaporation times (5s, 10s, and 15s). The quality of developed fingerprints was quantitatively assessed using parameters such as contrast, brightness, signal-to-noise ratio, and minutiae visibility. The results indicate that a gold evaporation time of 10s yielded the most effective development for both surface types. These findings highlight the importance of optimizing material-specific VMD parameters to enhance fingerprint visualization. Future studies should explore a wider range of substrates and experimental conditions to further refine the VMD process