Image retrieval method based on deep learning and Hash

Abstract

The invention relates to an image retrieval method based on deep learning and Hash. According to the image retrieval method, on the basis of powerful learning capacity of a deep convolutional neural network, deep features of images are extracted, and the problems of weak feature expression capacity and low retrieval precision caused by use of lower features of the images in the prior art are solved; a Hash layer is introduced for construction of a Hash function, learning of the deep features of the images and the construction of the Hash function are completed in the same process, an internal relation of the image features and the Hash function is explored, and the accuracy rate of the image retrieval is greatly increased; quantization error loss is added to a loss layer of the deep convolutional neural network, the expression capacity of Hash codes is enhanced, by means of a Softmax classifier loss module and a quantization error loss module, quantization errors caused by binaryzation in the Hash function are effectively reduced, and the accuracy rate of the image retrieval is further increased.

Description

technical field [0001] The invention relates to the field of image retrieval, in particular to an image retrieval method based on deep learning and hashing. Background technique [0002] With the advent of the era of big data, Internet image resources are growing rapidly, and how to quickly and effectively retrieve large-scale image resources to meet user needs needs to be solved urgently. In order to perform fast and efficient retrieval in large-scale image collections, hashing technology maps original images to binary hash codes while maintaining similarity. Due to the efficiency of binary hash codes in Hamming distance calculation and the advantages of storage space, hash codes are very efficient in large-scale image retrieval. Content-based image retrieval realizes the content expression of images by extracting the underlying visual features of images. Compared with these underlying features, the deep convolutional neural network is more able to obtain the intrinsic fe...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies in the prior art, the present invention provides an image retrieval method based on deep learning and hashing, and simultaneously learns image features and a deep convolutional network architecture for hash coding, adding a hash layer and continuous Quantization error loss caused by quantization of values ​​into binary codes, using this architecture to generate hash codes with very strong expressive capabilities, greatly improving the accuracy of image retrieval, and solving the problems caused by the weak expressive capabilities of generated hash codes in the prior art Problems with low retrieval accuracy

Images