Method for human shielded contour detection based on rotational depth learning

Abstract

The invention discloses a method for human shielded contour detection based on rotational depth learning. Firstly, the initial segmentation of an image is obtained by dividing and merging an input person image, and a segmentation contour of a target is extracted by regional merging of the color and the content; secondly, according to a rotation angle set, the image is rotated, sampled and labeledto obtain an edge image block set, based on a convolutional neural network, a depth model of the edge orientation detection is constructed, and a rotation image block acquisition set is used to traina shallow layer model and a deep layer model; and finally, the trained depth model of the edge orientation detection is used to detect the local contour orientation, the local contour orientation is evaluated for consistency, and the character segmentation contour orientation is extracted.

Description

technical field [0001] The invention belongs to the field of character occlusion contour detection, and more specifically, relates to a character occlusion contour detection method based on rotation deep learning. Background technique [0002] Contour detection refers to the process of using certain technology to extract the target contour, while robustly dealing with the influence of background noise and internal texture of the target. It is an important basis for techniques such as shape analysis, object detection, object recognition, and object tracking. At present, there are two types of contour detection methods, one is to use the traditional edge detection operator to detect the target contour, and the other is to extract the usable mathematical model from the human visual system to complete the target contour detection. [0003] Contour detection based on edge detection is a relatively common method. It mainly defines the low-level mutation of features such as bright...

AI Technical Summary

Problems solved by technology

The above method can get a clear object outline map very well, but there is no occlusion orientation of the outline. Therefore, it is impossible to effectively judge whether the area corresponding to the outline is a foreground area, which is not conducive to subsequent person detection and analysis.

Images