introduction

orange-infected-wound-trained-model is a supervised trained Neural Network model, trained to classify images of wound as infected wound or clean wound. This trained model is to be use within Orange Data Mining program.

requirements

  1. Orange Data Mining
  2. Image Analytics Addon for Orange Data Mining

usage

Open Orange Data Mining program, then drag “Import Images”, “Image Embedding”, “Load Model”, “Predictions” and “Image Viewer” elements into canvas and connect the elements in these particular order

Then double click “Import Images” and select the folder containing the wound images you want to classify.

Double click “Image Embedding” and select “SqueezeNet (local)” for the Embedder

Now its time to load the trained model. Download “wound nn.pkcls” from this repository then double click “Load Model”. Browse for the file that you’ve just downloaded

Orange Data Mining program will now evaluate the embedded wound images that you’ve selected. Double click on “Predictions”. Predictions will show you the neural network prediction confidence. In this case, the value for infected and clean wound is 1 (means full confidence)

Double click on Image Viewer to view the images. Select “Neural Network” from “Title Attribute” to view the Neural Network result. Notice the Neural Network result is displayed under the images.

license

Copyright (C) 2019 Mohd Kholid Yaacob (http://mrharmonies.blogspot.com)

This source is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

This code is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

A copy of the GNU General Public License is available on the World Wide Web at http://www.gnu.org/copyleft/gpl.html. You can also obtain it by writing to the Free Software Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston, MA 02110-1335, USA.