A skincare system

Abstract

The present invention discloses a system (100) for detecting human skin surfaces comprising an image acquisition module (110), configured for capturing a plurality of images of a patient; an image pre-processing module (120), configured for converting the captured images into an RGB image format and resize the captured images to a standardized size; a skin detection module (130), configured for annotating one or more skin regions and one or more non-skin regions within the captured images; and an image segmentation module (140), configured for employing a convolutional neural network to perform semantic segmentation on the captured images; wherein the image segmentation module (140) integrates an iterative optimization technique to fine-tune the convolutional neural network by optimizing a plurality of hyperparameters and architecture parameters, defining the architecture, training process, and performance of the convolutional neural network. (The most illustrative figure: FIG. 1)