Integrated interactive segmentation with spatial constraint for digital image analysis

摘要

An integrated interactive segmentation with spatial constraint method utilizes a combination of several of the most popular online learning algorithms into one and implements a spatial constraint which defines a valid mask local to the user's given marks. Additionally, both supervised learning and statistical analysis are integrated, which are able to compensate each other. Once prediction and activation are obtained, pixel-wised multiplication is conducted to fully indicate how likely each pixel belongs to the foreground or background.

主权项

1. A method of image processing programmed in a memory of a device comprising:
a. utilizing user input to generate a learned model for prediction; b. utilizing statistical analysis for activation; c. defining a local mask from the user input, wherein the local mask limits the image processing to a local area, further wherein the user input includes marking foreground and background in different colors; d. combining the prediction and the activation using pixel-wise multiplication, wherein the prediction and the activation are each probability maps indicating a percentage of a likelihood each pixel is in the foreground; e. performing object segmentation using a combination of the prediction and the activation.