Automated region segmentation on cartridge case base

Sakarya, Ufuk; Topcu, Osman; Leloglu, Ugur Murat; Soysal, Medeni; Tunalı, Erol

doi:10.1016/j.forsciint.2012.07.004

Cited by 8 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These systems automate the imaging of samples, identify regions of interest (ROI), and provide a numeric estimate of likely matches. They typically include an optical microscope, a signature analysis station, and a matching algorithm for an end-to-end ballistics evaluation [2,4,7,[13][14][15].…”

Section: Introductionmentioning

confidence: 99%

“…However, this manual method is not ideal because rapid processing becomes unrealistic with increasing numbers of cartridge case samples. Second, manual segmentation by a human operator injects bias and variation into the samples [14,15]. In other studies, the authors report using preprocessing steps to segment the images and separate the different types of marks.…”

Section: Introductionmentioning

confidence: 99%

“…This combination retains the true shape of the firing pin mark since this shape yields an identification value [14]. A few studies also focused predominantly on the automated segmentation of ROI on cartridge cases images using traditional image‐processing methods: a combination of methods including Connected Components Labeling, the Canny operator, and the Randomized Hough Transform; a second method using the image focus information for each pixel paired with a Hough transform [20], and thresholding technique using Surface Height Image information [15].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automated detection of regions of interest in cartridge case images using deep learning

Le Bouthillier,

Hrynkiw,

Beauchamp

et al. 2023

Journal of Forensic Sciences

View full text Add to dashboard Cite

This paper explores a deep‐learning approach to evaluate the position of circular delimiters in cartridge case images. These delimiters define two regions of interest (ROI), corresponding to the breech face and the firing pin impressions, and are placed manually or by an image‐processing algorithm. This positioning bears a significant impact on the performance of the image‐matching algorithms for firearm identification, and an automated evaluation method would be beneficial to any computerized system. Our contribution consists in optimizing and training U‐Net segmentation models from digital images of cartridge cases, intending to locate ROIs automatically. For the experiments, we used high‐resolution 2D images from 1195 samples of cartridge cases fired by different 9MM firearms. Our results show that the segmentation models, trained on augmented data sets, exhibit a performance of 95.6% IoU (Intersection over Union) and 99.3% DC (Dice Coefficient) with a loss of 0.014 for the breech face images; and a performance of 95.9% IoU and 99.5% DC with a loss of 0.011 for the firing pin images. We observed that the natural shapes of predicted circles reduce the performance of segmentation models compared with perfect circles on ground truth masks suggesting that our method provide a more accurate segmentation of the real ROI shape. In practice, we believe that these results could be useful for firearms identification. In future work, the predictions may be used to evaluate the quality of delimiters on specimens in a database, or they could determine the region of interest on a cartridge case image.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated detection of regions of interest in cartridge case images using deep learning

Le Bouthillier,

Hrynkiw,

Beauchamp

et al. 2023

Journal of Forensic Sciences

View full text Add to dashboard Cite

show abstract

“…In addition to obtaining the surface topography, another important step in the acquisition of cartridge case data involves segmentation of the area of marks on the base of a cartridge case. BALİSTİKA 2010 performs this semisupervised . In this study, the firing pin impression, trace of primers, and the combination of these areas formed on the cartridge were taken into account to compare the similarities and differences through correlation ranking method using BALISTIKA 2010 for selected pistol makes.…”

Section: Introductionmentioning

confidence: 99%

Investigation of Ballistic Evidence through an Automatic Image Analysis and Identification System

Kara

2016

Journal of Forensic Sciences

View full text Add to dashboard Cite

Automated firearms identification (AFI) systems contribute to shedding light on criminal events by comparison between different pieces of evidence on cartridge cases and bullets and by matching similar ones that were fired from the same firearm. Ballistic evidence can be rapidly analyzed and classified by means of an automatic image analysis and identification system. In addition, it can be used to narrow the range of possible matching evidence. In this study conducted on the cartridges ejected from the examined pistol, three imaging areas, namely the firing pin impression, capsule traces, and the intersection of these traces, were compared automatically using the image analysis and identification system through the correlation ranking method to determine the numeric values that indicate the significance of the similarities. These numerical features that signify the similarities and differences between pistol makes and models can be used in groupings to make a distinction between makes and models of pistols.

show abstract

“…Uma estratégia para comparação seria a segmentação da base do estojo em áreas de interesse e o uso de métodos baseados em modelos para automatização das comparações (SAKARYA et al, 2012 Os estudos citados demonstram que os desafios nesta área são em duas frentes, captura da imagem com extração de informações relevantes que possam funcionar como uma assinatura daquele elemento de munição, e subsequente estabelecimento de técnica para comparação de suas assinaturas. i é o tamanho do banco de dados (ver Tabela 1); e P (n) é a probabilidade cumulativa de um acerto até a posição n.…”

Section: Técnicas De Captura De Imagens E Correlação Automatizadaunclassified

Avaliação de parâmetros que afetam a efetividade da comparação balística automatizada

Santos¹

View full text Add to dashboard Cite

Automated region segmentation on cartridge case base

Cited by 8 publications

References 16 publications

Automated detection of regions of interest in cartridge case images using deep learning

Automated detection of regions of interest in cartridge case images using deep learning

Investigation of Ballistic Evidence through an Automatic Image Analysis and Identification System

Avaliação de parâmetros que afetam a efetividade da comparação balística automatizada

Contact Info

Product

Resources

About