Skip to content

Latest commit

 

History

History
18 lines (12 loc) · 1.34 KB

File metadata and controls

18 lines (12 loc) · 1.34 KB

Deep-Learning-Based-On-Device-Breast-Cancer-Classification-using-Thermography

An android application for classification of thermal images of Breast using MobileNet

Dataset is acquired from Database for Mastology Research with Infrared Image(DMR-IR). Dataset contains IR images and thermal maps taken in two protocols:-

  1. Dynamic Infrared Thermography (DIT)
  2. Static Infrared Thermography (SIT)

Both protocols contains

  1. Grayscale image
  2. Thermal Maps

In this work Thermal maps of both protocols are used. Thermal maps are similar to heat maps, they contain arrays of temperature at a given point in the image.

Experimentation with 13 models was done with fine-tunning. Comparions of the models is given here

Taking size and accuracy into conideration, an android application was made. Screenshots of the Android Application are given below-

image image image