National Repository of Grey Literature 9 records found  Search took 0.00 seconds. 
Detection of persons and evaluation of gender and age in image data
Dobiš, Lukáš ; Vičar, Tomáš (referee) ; Kolář, Radim (advisor)
Táto diplomová práca sa venuje automatickému rozpoznávaniu ludí v obrazových dátach s využitím konvolučných neurónových sieti na určenie polohy tváre a následnej analýze získaných dát. Výsledkom analýzy tváre je určenie pohlavia, emócie a veku osoby. Práca obsahuje popis použitých architektúr konvolučných sietí pre každú podúlohu. Sieť na odhad veku má natrénované nové váhy, ktoré sú vzápätí zmrazené a majú do svojej architektúry vložené LSTM vrstvy. Tieto vrstvy sú samostatne dotrénované a testované na novom datasete vytvorenom pre tento účel. Výsledky testov ukazujú zlepšenie predikcie veku. Riešenie pre rýchlu, robustnú a modulárnu detekciu tváre a ďalších ludských rysov z jedného obrazu alebo videa je prezentované ako kombinácia prepojených konvolučných sietí. Tieto sú implementované v podobe skriptu a následne vysvetlené. Ich rýchlosť je dostatočná pre ďalšie dodatočné analýzy tváre na živých obrazových dátach.
Age estimation from retinal images
Kadlec, Vojtěch ; Jakubíček, Roman (referee) ; Kolář, Radim (advisor)
Human age is considered an important biometric parameter that is often difficult to determine. Previous studies have shown that the non-specific general anatomical and physiological characteristics seen on fundus images are all likely signs of ageing. This bachelor thesis focuses on age estimation from retinal images. The first part of the thesis deals with the retina of the human eye as such, including physiological and pathological changes during aging. Then the principles of neural networks and principles of age classification from image data are described. The practical part is devoted to the implementation of the algorithm itself. The convolutional neural network framework is based on the ResNet-34 architecture together with the Consistent Rank Logits algorithm estimating age as an ordinal variable. The best model achieved a mean absolute error of 3.42 years, outperforming existing models estimating age from retinal images. All algorithms are implemented in the Python programming language using the Pytorch library.
Relationship between the age of freshwater pearl mussel and the number of increments in shell section
TOMAN, Jan
The aim of this thesis was to determine the relationship between the age of dead individuals of the freshwater pearl mussel (Margaritifera margaritifera) and the number of incremental lines on the shell section. A total of 124 pearl mussel shells of known age were collected from several Czech, German and Finnish localities. There was a statistically significant difference between the actual age and the age according to the number of incremental lines on shell section. On average, the age determined by increment lines was higher than the actual age by 2.6 years, which is 42 ? 25% (? S.D.). Thus, the increment lines on the section of the pearl mussel shell do not correspond to the annual increments. It was confirmed that increment lines count on the shell surface or estimating age from shell size are less reliable approaches to determine the true age of individuals.
Geometric-morphometric approach to age and sex variability of the acetabulum
Cibulková, Simona ; Brůžek, Jaroslav (advisor) ; Bejdová, Šárka (referee)
6 Abstract This thesis compared age and sex differences in lunate surface morphology using a 3D geometric-morphometric approach. The acetabulum of 240 individuals was compared using landmarks and semilandmarks placed along the edge of the lunate surface. The individuals ranged in age from 20 to 90 and came from three geographic areas. This thesis was based on the study of San-Millán et al. (2017a) that used a 2D geometric-morphometric approach to investigate the shape of the acetabulum. Analyses in this thesis showed that size, sex, and age significantly affect the acetabular shape. The differences between both sexes can be observed in the size and depth of the acetabulum, the width of the acetabular notch, and the amount of bone growth at the acetabular horns and along the edges of the lunate surface. Both sexes exhibit age- related changes, which are linked to gradual deposits of bone along the edge of the lunate surface, the acetabular horns, and the acetabular fossa, which tends to lose the 3-lobed cloverleaf shape. According to the geometric-morphometric analysis conducted in this thesis, the acetabulum provides more accurate age estimates for individuals younger than 65 years of age. Keywords: Bioarchaeology, forensic anthropology, age estimation, sex estimation, acetabulum, lunate surface of hip...
Age estimation from human biological material based on DNA methylation
Matoušková, Laura ; Kulichová, Iva (advisor) ; Priehodová, Edita (referee)
DNA methylation is one of the epigenetics modifications which performs many functions in human body by silencing genes and maintaining genomic stability. Changes in DNA methylation accumulate during life and current studies indicate that the level of methylation in certain loci (CpG dinucletotides) correlates with chronological age. However, the methylation pattern changes not only with increasing age but it can also be affected by our lifestyle choices and biogeographical origin. Therefore, the key for succesful age estimation is finding such CpG loci that are independent (or as little dependent as possible) on other variables. This bachelor thesis describes the issue of age estimation based on DNA analysis of specific methylated DNA loci with focus on its application in forensic practice.
Detection of persons and evaluation of gender and age in image data
Dobiš, Lukáš ; Vičar, Tomáš (referee) ; Kolář, Radim (advisor)
Táto diplomová práca sa venuje automatickému rozpoznávaniu ludí v obrazových dátach s využitím konvolučných neurónových sieti na určenie polohy tváre a následnej analýze získaných dát. Výsledkom analýzy tváre je určenie pohlavia, emócie a veku osoby. Práca obsahuje popis použitých architektúr konvolučných sietí pre každú podúlohu. Sieť na odhad veku má natrénované nové váhy, ktoré sú vzápätí zmrazené a majú do svojej architektúry vložené LSTM vrstvy. Tieto vrstvy sú samostatne dotrénované a testované na novom datasete vytvorenom pre tento účel. Výsledky testov ukazujú zlepšenie predikcie veku. Riešenie pre rýchlu, robustnú a modulárnu detekciu tváre a ďalších ludských rysov z jedného obrazu alebo videa je prezentované ako kombinácia prepojených konvolučných sietí. Tieto sú implementované v podobe skriptu a následne vysvetlené. Ich rýchlosť je dostatočná pre ďalšie dodatočné analýzy tváre na živých obrazových dátach.
Age estimation in two recent European population by assessment of permanent tooth mineralization
Černá, Magdaléna ; Velemínská, Jana (advisor) ; Stránská, Petra (referee)
Age estimation is a common requirement in forensic, bioarcheological and biomedical practice. This master thesis deals with age estimation based on permanent tooth mineralization according to Demirjian et al. (1973). The research material consisted of orthopantomograms of 716 Czech and 743 French children aged between 4 and 15 years. The purpose of this study was to analyse the suitability of the original French-Canadian standards for age estimation (Demirjian a Goldstein, 1976) and the recent Belgian standards (Willems et al., 2001) in Czech and French population. Another aim of the study was to evaluate the accuracy of the neural network method that represents a completely new approach in data prediction. In order to express the accuracy of estimate we used mean and median of difference between chronological and dental age, and RMS error. Using logistic regression, differences in tooth mineralization between Czech and French population and between girls and boys were also evaluated. Our results indicate that the French-Canadian standards gave a consistent overestimation of dental age compared with chronological age. Mean difference was 0,33 years for Czech children and 0,45 and 0,46 years for French girls and boys, respectively. We found that Willem's method and neural network method were more...
Age estimation based on human dentice
Černá, Magdaléna ; Velemínská, Jana (advisor) ; Brůžek, Jaroslav (referee)
Age estimation plays a very important role not only in forensic and biomedicinal anthropology, but also in other disciplines, such as bioarcheologie or paleoanthropology. Teeth are considered the most reliable age indicators during childhood and adolescence. Teeth are resistant to the effects of external and internal factors, and because of that they decay a long time after a death of an individual. While age estimation in children and adolescents can be based on mineralization or eruption of the teeth, in adults it is focused on retrogressive changes in dentition. There have been established several methods and their modification for age assessment based on these criteria. This thesis aims to generally summarize tthe principles and methods of age estimation based on dentition both in immature and adult individuals.
Dental enamel hypoplasia in early medieval population of Rajhrad.
Zahradníková, Mariana ; Trefný, Pavel (advisor) ; Stránská, Petra (referee)
The objective of this study was the assessment of linear enamel hypoplasia (LEH) in early medieval Great Moravian population sample - Rajhrad. Linear enamel hypoplasia represents the disruption of enamel matrix secretion during the growth of the tooth crown, which is related to a generalized growth disturbance. This is why it is considered as a nonspecific stress marker. The incidence of LEH could reflect stress factors during the life of early medieval population. The aim of this study was to asses the frequency and timing of the LEH. The incidence of LEH was supposed to be high because of poorer nature of this cemetery. The results of our study could confirm or falsify this assumption and determine relation between LEH and socio-economic status. The timing of LEH was estimated from regression equations consisting of distance from LEH to CEJ (cemento-enamel junction) and crown height of upper and lower canines. 108 individuals from approximately 4 - 15 years were observed. The frequency was high according to the assumption - 88 %. That confirms poorer life conditions. The range of mean age of LEH formation was from 2,94 - 4,72 years in individuals with multiple LEH incidence. The mean age of single LEH formation was approximately 3,98 years. The earliest onset of LEH in the pooled sample occurred...

Interested in being notified about new results for this query?
Subscribe to the RSS feed.