Image visual semantic segmentation method based on two-way region attention coding and decoding

Abstract

The invention discloses an image visual semantic segmentation method based on two-way region attention coding and decoding. The method comprises the following specific steps: acquiring an image sample of a specific scene in advance; normalizing an RBG channel of the sample image, and preparing to train a depth model; encoding an image through a two-way encoder to obtain multi-scale and refined image depth features; carrying out adaptive channel feature enhancement on different distributed targets through regional information by using a decoder based on regional attention; fusing encoder shallow layer features and decoder deep layer features in different extraction stages through skip-connection, and multiplexing the depth features to the maximum extent; and finally, mapping from a final convolutional layer of the deep neural network to an original image, and classifying each pixel point to obtain a final image visual segmentation image. The invention can be embedded into equipment such as a monitoring probe and the like, and the image with complicated distribution is guided through regional information, so that accurate visual semantic segmentation of the image is realized.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image visual semantic segmentation method based on two-way area attention coding and decoding. Background technique [0002] With the development of society, semantic segmentation, as an important technology of image visual semantic segmentation method based on two-way regional attention codec, has attracted more and more attention, and its application scope has gradually expanded. From conventional daily image semantic segmentation to Semantic segmentation extended to specific application scenarios. After deep learning is applied to semantic segmentation, the semantic segmentation of conventional images has been greatly improved, but the semantic segmentation effect for specific work scenes is not very obvious, because conventional images are taken with the ground as a reference, The target in the photo is horizontal and vertical, and occupies most of the area of ​​...

AI Technical Summary

Problems solved by technology

Method 1 and method 2 are limited by manpower and the angle of the camera, lack of pertinence, and cannot filter out most unnecessary images and deal with emergencies

[0004] In order to enhance the flexibility of the semantic segmentation system, researchers embed the semantic segmentation system into devices containing cameras such as surveillance probes, and then realize the semantic segmentation of images through the movement of the camera. However, this method usually uses computer vision The semantic segmentation task has replaced the carrying platform, ignoring the characteristics of the image itself

In practical application scenarios, this kind of method is often because the proportion of the target in the lens is too small, and the direction is unpredictable. At the same time, due to the angle of view of the camera, the target is easily blocked by other objects, resulting in the uncertainty of the boundary contour of different targets. Unable to achieve precise segmentation of the target

Images