Método automático para cálculo de assimetria térmica corporal em imagens infravermelhas
Thermography is a non-invasive examination technique that uses infrared images for verification and analysis of the body temperature variations. The exam usually evaluates the distribution of temperature in symmetrical regions. Despite providing important information for medical diagnosis, the proce...
| Autor principal: | Silva, Gabriel de Andrade |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2020
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/16003 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Thermography is a non-invasive examination technique that uses infrared images for verification and analysis of the body temperature variations. The exam usually evaluates the distribution of temperature in symmetrical regions. Despite providing important information for medical diagnosis, the process of selection of the regions to be analyzed is usually done manually. To automate this process of selection, it is possible to use the Viola and Jones algorithm, which allows the detection of objects and shapes. The objective of this work is to research and develop an automatic method for the detection of symmetrical regions of the frontal part of the human body’s lower limbs. Volunteers images collected from a camera sensitive to infrared radiation emitted by the body were used for this. The developed software uses the Viola and Jones algorithm to identify the knees in the collected images and from the knees detection, the software identifies the thighs and the legs. After the detections, the software automatically determines the minimum temperature, maximum temperature, average temperature and the temperature difference between the identified symmetrical regions. The final results of the development brought a methodology of detection capable of detecting and calculate automatically the temperature variations of the thighs, knees and legs, and compare the results with the medical exam protocol proposed by Uematsu. The method used to detect the knees in infrared images was able to correctly identify 87,5% of the knees in the images. |
|---|