Abstract:
The objective of this project is that the system will be able to analyse, detect and
classify skin diseases. This report explores image classification methods used for the
classification of five major types of skin diseases and discussed different stages
involved in image processing like the data augmentation stage that include image pre processing, segmentation of images and feature extraction. Finally write and
implement algorithms and code using python.
This project uses the Deep Convolutional Neural Network technique algorithms for
image processing. Deep Convolutional Neural Network provides features extraction
and detection and this is main advantage ofDCNN. Architecture ofdeep convolutional
neural network are discussed, different convolutional layers and their working is tested
by applying different filter sizes. Finalize our model after setting suitable number of
layers and filters and proper trials and error.
Collect image dataset from authentic site and apply pre-processing and proper data
augmentation. In the process data segmentation, filtering, resizing and features
extraction are also performed. Than determination of proper classification technique,
such as learning rate, batch size, number of epochs, and optimizer’s type with an
objective to made model better and better, and finally assessing the overall accuracy.
In this method, the inputs are images of a specific skin disease as object. This system
is specially created to help out dermatologists for accurate and quick predictions.
Recommendations for future development and conclusions are also included in the
report.
