Image retrieval method based on depth learning and semantic segmentation

Abstract

The invention discloses an image retrieval method based on depth learning and semantic segmentation, which comprises the following steps of: reading an image and preprocessing the image; encoding theimage into a set of characteristic graphs through depth learning by any convolution layer of the depth neural network; carrying out semantic segmentation on the image to obtain a category label of each pixel of the segmented image; carrying out weighting processing on the category labels according to each pixel category label on the characteristic graphs and the set category weight to obtain a setof weighted characteristic graphs; coding the set of weighted characteristic graphs to a feature vector of a fixed length, and carrying out normalization processing, and characterizing the final coded feature vector of the image by using a normalized characteristic vector; carrying out similarity calculation and returning the search result. According to the invention, the semantic segmentation technology is introduced into the feature code of image retrieval, and the retrieval effect is greatly improved. When the weight of each category of the image is acquired, the provided manual design method based on the prior knowledge and parameter learning method of the depth neural network are very effective.

Description

technical field [0001] The invention belongs to the field of image retrieval and relates to an image retrieval method based on deep learning and semantic segmentation. Background technique [0002] With the rapid development of Internet technology and the popularization of smart terminals, images have become the main way for people to record and share information, and image retrieval technology has emerged. Image retrieval is the technology of querying the input image content to retrieve similar images, and it is a search technology about graphic image information retrieval. [0003] The image feature representation is the connection between the pixel information of the image and human perception of things, and the image feature is the retrieval condition. [0004] Existing technologies generally use machine learning methods such as sift operator, fisher vector, or VLAD to extract features, and the extracted feature vector has a large dimension, resulting in high cost of da...

AI Technical Summary

Problems solved by technology

The existing CNN-based image retrieval algorithm does not take into account the enhancement of the salient area features of the image when encoding image features, such as retrieving buildings, the area where the building is located is the salient area, but this salient area The feature encoding of the region is the same as the encoding of its existing background (such as the sky, grass), which is obviously not reasonable

Images