Abstract
At this stage, due to the different methods of expressing attribute characteristics between cross-type images, the performance measurement accuracy is low, only approximately 78%. This article describes an image similarity measurement method across types, focusing on the image edge feature autoencoder and the edge feature map similarity measurement network. Taking advantage of the good effect of the autoencoder in feature extraction, an autoencoder that can efficiently extract edge feature maps is trained. In terms of similarity measurement, similar to twin neural networks, two deep network models are established for similarity measurement, and image data classification training is used instead of the training method used in traditional twin neural networks. The original image is imported into the autoencoder to obtain the edge feature map of the original map, and then the resulting edge feature map is imported into the similarity measurement network. The similarity measurement is carried out with the line diagram in the library in which the similarity measurement accuracy of cross-type images can reach 92%.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Akbas, C., Günay, O., Tasdemir, K., Cetin, A.: Energy efficient cosine similarity measures according to a convex cost function. Signal Image Video Process. 11(2), 66 (2017)
Cao, Y., Wu, D., Duan, Y.: A new image edge detection algorithm based on improved Canny. J. Comput. Methods Sci. Eng. 20(2), 629–42 (2020)
Chen, T., Kornblith, S.N.: A simple framework for contrastive learning of visual representations. In: International Conference on Machine Learning (2020)
Jiang, F., Grigorev, A., Rho, S., Tian, Z., Fu, Y., Jifara, W., Adil, K., Liu, S.: Medical image semantic segmentation based on deep learning. Neural Comput. Appl. 29(5), 1257–65 (2018)
Nair, V., Hinton, G.E.: Rectified linear units improve restricted Boltzmann machines. In: ICML (2010)
Niizumi, D., Takeuchi, D., Ohishi, Y., Harada, N., Kashino, K.: BYOL for audio: self-supervised learning for general-purpose audio representation (2021)
Shi, W., Liu, S., Jiang, F., Zhao, D., Tian, Z.: Anchored neighborhood deep network for single-image super-resolution. EURASIP J. Image Video Process. 1, 1–2 (2018)
Sulistyo, W.Y., Riadi, I., Yudhana, A.: Comparative analysis of image quality values on edge detection methods. Jurnal RESTI Rekayasa Sistem dan Teknologi Informasi 4(2), 345–51 (2020)
Wang, F., Zuo, W., Lin, L., Zhang, D., Zhang, L.: Joint learning of single-image and cross-image representations for person re-identification. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)
Wang, G., Wang, S., Chi, W., Liu, S., Fan, D.A.: person reidentification algorithm based on improved Siamese network and hard sample. Math. Probl. Eng. 6, 66 (2020)
Wei, Z., Zhang, X.: Feature extraction and retrieval of ecommerce product images based on image processing. Traitement du Signal 38, 66 (2021)
Xu, Z., Ji, X., Wang, M., Sun, X.: Edge detection algorithm of medical image based on canny operator. J. Phys. Conf. Ser. 6, 66 (2021)
Zagoruyko, S., Komodakis, N.: Learning to compare image patches via convolutional neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2015)
Zhang, X., Chen, F.: Lane line edge detection based on improved adaptive canny algorithm. J. Phys. Conf. 6, 66 (2020)
Zhang, K., Liao, Q.: FPGA implementation of eight-direction Sobel edge detection algorithm based on adaptive threshold. J. Phys. Conf. Ser. 6, 66 (2020)
Zhu Qianqian, W.D.: A SVM recommendation IoT model based on similarity evaluation and collaborative filtering of multi-angle knowledge units. Int. J. Comput. Appl. 42(3), 278–81 (2020)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Yue, J., Jiang, B., Fu, J. et al. Cross-type image similarity measurement. SIViP 17, 1227–1234 (2023). https://doi.org/10.1007/s11760-022-02330-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-022-02330-z