Intelligent acne spot removing method and system for portrait photo

Abstract

The invention discloses an intelligent acne spot removing method and system for a portrait photo, and belongs to the field of PC image processing software, and the removing method comprises the following specific steps: (1) receiving an original picture of a user for segmentation detection; (2) performing connected domain marking on the feature map so as to perform judgment and screening; and (3) constructing an acne spot removal network, and carrying out data fusion on the image. The skin segmentation and acne spot classification network is constructed by using the convolutional neural network, the acne spot judgment module can be accurately combined, and the acne spot judgment accuracy and the acne spot removal effect can be improved.

Description

technical field [0001] The invention relates to the field of PC image processing software, in particular to a method and system for intelligently removing pimples from portrait photos. Background technique [0002] At present, most of the portrait skin acne and speckle removal functions in mobile or PC-side image processing software are manual speckle removal or acne spot detection based on traditional image algorithms and acne spot removal similar to inpainting, and a few of them use deep learning-based acne spot detection Acne spot removal method, the manual spot removal method requires manual operation by the user, which is complicated and time-consuming; the traditional image method for acne spot removal is inaccurate, and there are a large number of false detections and missed detections; the method based on deep learning acne spot detection, due to the size It is relatively small, and it is easy to cause missed detection, and the overall effect is poor; at present, the...

AI Technical Summary

Problems solved by technology

[0002] At present, most of the portrait skin acne and speckle removal functions in mobile or PC-side image processing software are manual speckle removal or acne spot detection based on traditional image algorithms and acne spot removal similar to inpainting, and a few of them use deep learning-based acne spot detection Acne spot removal method, the manual spot removal method requires manual operation by the user, which is complicated and time-consuming; the traditional image method for acne spot removal is inaccurate, and there are a large number of false detections and missed detections; the method based on deep learning acne spot detection, due to the size It is relatively small, and it is easy to cause missed detection, and the overall effect is poor; at present, these methods cannot meet the needs of high-definition image retouching with increasing resolution, especially in industries such as studio intensive retouching and e-commerce model photo beautification; therefore, It is particularly important to invent a method and system for intelligent acne spot removal in portrait photos;

[0003] The intelligent acne spot removal method and system for portrait photos have low accuracy rate and acne spot removal effect, and it is impossible to combine accurate acne spot judgment modules; therefore, we propose an intelligent acne spot removal method and system for portrait photos

Images