Textile image fingerprint retrieval method

Abstract

The invention discloses a textile image fingerprint retrieval method. The method comprises the steps of image fingerprint acquisition, image index construction and image retrieval. The image fingerprint acquisition comprises the steps of acquiring fingerprints of a to-be-detected textile image and all textile images in a textile image library, defining a plurality of repeated local fine-grained features in the textile images as fingerprint parts of the textile images, and defining image features generated after the extracted fingerprint parts are fused as the fingerprints of the textile images; taking the fingerprint of each image in the image library as the feature of the corresponding image and correspondingly storing same with the image name; and calculating the distance between the fingerprint of the image to be detected and the image fingerprint features in the image library by adopting a Cosine cosine distance, and taking out the picture with the closest distance, namely the retrieved textile image. According to the invention, the user can conveniently and quickly find the textile image using the fingerprint component, and the retrieval efficiency is improved. Precise image retrieval is achieved through fingerprints, and the accuracy of textile image fingerprint retrieval is improved.

Description

technical field [0001] The invention belongs to the technical fields of target detection and image retrieval, and relates to a textile image fingerprint retrieval method, in particular to a local fine-grained feature extraction method of textile images. Background technique [0002] In recent years, with the development of the textile image design industry, the number of textile images has grown rapidly, showing an exponential growth trend. How to better apply textile image data to achieve convenient, fast and accurate query and retrieval of image information required by users, so that managers can be liberated from a large number of monotonous manual management work, has become an urgent problem that people need to solve question. Image retrieval technology has a wide range of application scenarios in various industrial fields. There are problems such as time-consuming and laborious classification and search of textile images, and low image accuracy required for retrieval ...

AI Technical Summary

Problems solved by technology

Fine-tuning the model is very important for learning fine-grained image classification in image retrieval, but the fine-tuning model will lead to the problem of "catastrophic forgetting", almost losing the recognition ability of the original data set, and the fine-tuning method cannot transfer the model to different data sets

Images