Data-Driven Color Augmentation Techniques for Deep Skin Image Analysis

By Adrian Galdran, Aitor Alvarez-Gila, Maria Inês Meyer, C. L. Saratxaga, Teresa Araújo, Estíbaliz Garrote, Guilherme Aresta, Pedro Costa, Ana Maria Mendonça, Aurélio J. C. Campilho. 2017

68 citations recorded by the source corpus.

Dermoscopic skin images are often obtained with different imaging devices, under varying acquisition conditions. In this work, instead of attempting to perform intensity and color normalization, we propose to leverage computational color constancy techniques to build an artificial data augmentation technique suitable for this kind of images. Specifically, we apply the \emph{shades of gray} color constancy technique to color-normalize the entire training set of images, while retaining the estimated illuminants. We then draw one sample from the distribution of training set illuminants and apply it on the normalized image. We employ this technique for training two deep convolutional neural networks for the tasks of skin lesion segmentation and skin lesion classification, in the context of the ISIC 2017 challenge and without using any external dermatologic image set. Our results on the validation set are promising, and will be supplemented with extended results on the hidden test set when available.

Subfields: Atomic and Molecular Physics, and Optics, Cell Biology, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Style transfer-based image synthesis as an efficient regularization technique in deep learning (2019, methodology): “The category of affine transformation methods covers techniques that use linear transformations to the image, such as rotation, shear, reflection, scaling [10], [13], while the most popular geometric transformations are hue modification, contrast modification, white-balance, sharpening [14].”
A comprehensive survey of recent trends in deep learning for digital images augmentation (2021, unknown): “(Galdran, et al. 2017).”
A systematic review on overfitting control in shallow and deep neural networks (2021, background): “In some instances (Kwasigroch et al. 2017; Wąsowicz et al. 2017; Galdran et al. 2017), the experimental results of the affine transition augmentation have been presented.”
An analysis of rotation matrix and colour constancy data augmentation in classifying images of animals (2018, methodology): “The research by Galdran et al. (2017) proposed a DA method adapted for skin lesion analysis with neural networks with emphasis on the use of colour constancy to normalize the colour information of images within a training set.”