Image processing method and device, electronic device and computer readable storage medium

Abstract

The invention discloses an image processing method and device and an electronic device. The method comprises the steps that a depth image of a current user is acquired, and a three-dimensional background image of a scene where the current user is located is acquired; edge extraction is performed on the three-dimensional background image to acquire depth data, in the three-dimensional background image, of edge pixels of a target object in the three-dimensional background image; according to the depth image of the current user and the depth data of the edge pixels of the target object, whether the current user collides with the target object in the scene is judged; and if yes, audio of a sound which should be made by the target object in collision is interposed. According to the embodiment,a person region and a target object region extracted through the depth image are more accurate, and particularly edge pixels of the person region and edge pixels of the target object region can be accurately calibrated. Besides, when a person collides with the target object, image processing is further performed to simulate the sound which should be made by the target object in collision, and therefore user experience is improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, and in particular to an image processing method, device, electronic device and computer-readable storage medium. Background technique [0002] The existing technologies for reminding people of collisions with objects in the background usually use feature points to extract the outlines of people and objects, and judge whether the current user collides with objects in the background according to the outlines of people and objects. However, the accuracy of character outlines and object outlines extracted using feature points is not high, especially the boundaries of characters and objects cannot be accurately marked, which affects the judgment of whether characters and objects collide. In addition, in the prior art, when it is determined that a person collides with an object, the user can only be informed through a simple reminder, and no further processing is performed on the colli...

AI Technical Summary

Problems solved by technology

However, the accuracy of character outlines and object outlines extracted using feature points is not high, especially the boundaries between characters and objects cannot be accurately calibrated, which affects the judgment of whether characters and objects collide

In addition, in the prior art, when it is determined that a person collides with an object, the user can only be informed through a simple reminder, and the collision event is not further processed.

Images