Skip to content

A comparative study for skin lesion segmentation and melanoma detection where deep learning methods can perform very well without complex pre-processing techniques except for normalization and augmentation.

Notifications You must be signed in to change notification settings

nahin333/A-Comparative-Study-of-Neural-Network-Architectures-for-Lesion-Segmentation-and-Melanoma-Detection

Repository files navigation

Deep learning applied with augmentation can save us from complex pre-processing steps and it studies the performance of different neural network architectures for lesion segmentation as an integral part of image processing-based techniques and finally evaluates network architectures for melanoma detection on dermoscopic images using transfer learning and presents a comparative view of those studies.

Screenshot (2595)

About

A comparative study for skin lesion segmentation and melanoma detection where deep learning methods can perform very well without complex pre-processing techniques except for normalization and augmentation.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published