Skin lesion classification from dermoscopy images using color and shape features

Journal Title: روش های هوشمند در صنعت برق - Year 2017, Vol 8, Issue 29

Abstract

In this study, a new algorithm for dermoscopy image classification into two types of malignant and benign is presented. At first, one preprocessing step to remove noise and also enhance image quality is performed. After that using Otsu thresholding, the lesion is separated from the healthy area. Then color and shape features are extracted from the segmented image. The colored features based on statistical moments of quantized grayscale and quantized color histogram are defined. These features demonastrare distribution of color components. Moreover, the shape features are extracted information of the segmented regions with two scenarios. In the first scenario, the features are represented the expantion of region and in the second scenario, the features are represented the edge variations of the extracted regions. Finally, the classification procedure is performed using K-Neasert Neighbor (KNN), Decision Tree, Support Vector Machine (SVM) and Adaboost. The proposed algorithm is evaluated on a standard database consisting of 200 images. The results show that classification using Adaboost classifier obtains the precision rate, accuracy rate and sensitivity rate of 96%, 96.7% and 95%, respectively.

Authors and Affiliations

Hamid reza Javadi, Hossein Pourghassem

Keywords

Related Articles

Design and Fabrication of Novel Single- and Dual-band Bandpass Filters with Modern Zero-Degree Feed Structure for Wireless Communications

In this paper, there are three novel bandpass filters (BPFs), consist of single- and dual-band, presented in a same size, which all the filters, have been designed, optimized and finally fabricated and measured, based on...

Design and Simulation of a Wilkinson Power Divider with High Isolation for Tri-Band Operation Using PSO Algorithm

In this article, a microstrip tri-band Wilkinson power divider, using the equivalent three section transmission line instead of common quarter-wavelength transmission line with help of PSO algorithm, design and simulated...

Design and Simulation of a New Optimized Full-Adder Using Carbon Nano Tube Technology

The full adder circuit is one of the most significant and prominent fundamental parts in digital processors and integrated circuits since it can be used for implementing all four basic computational functions including:...

Neuro-adaptive control in beating heart surgery based on the viscoelastic tissue model

In this paper, the problem of 3D heart motion in beating heart surgery is resolved by proposing a parallel force-motion controller. Motion controller is designed based on neuro-adaptive approach to compensate 3D heart mo...

Design and Implementation of Half-Bridge LLC Resonant Converter by FHA Technique

Although in existence for many years, only recently has the LLC resonant converters, in particular in its half-bridge implementation, gained in the popularity it certainly deserve. The advantages such as high efficiency,...

Download PDF file
  • EP ID EP317652
  • DOI -
  • Views 117
  • Downloads 0

How To Cite

Hamid reza Javadi, Hossein Pourghassem (2017). Skin lesion classification from dermoscopy images using color and shape features. روش های هوشمند در صنعت برق, 8(29), 33-40. https://europub.co.uk/articles/-A-317652