Computer vision for larval structures identification applied to forensic science.
No hay miniatura disponible
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
Descripción
The diptera maggots are used in forensic entomology to estimate the post-mortem interval (PMI). Maggots have a wide range of morphological and structural features that aid in the identification. In order to assist in the necrophagous larvae identification, this research aims to develop a software using computer vision and machine learning to automate the classification process. Diptera maggots were collected in a dead pig at the capital of Mato Grosso do Sul state, Campo Grande. The maggots were identified and photographed at a light microscope (5x objective). Next, the images were processed, the features extraction was performed using an extractor in Python language. The classification of the images were tested with AdaBoost, Random Forest, Random Tree and SMO classifiers. The SMO the best performa.
Palabras clave
Entomology, Insect larvae, Computer vision