Electronics Optics & Control, Volume. 30, Issue 11, -1(2023)
A Personalized Expert Recognition Algorithm for Long-Tailed Images
In practical applications of image recognition, the training data tend to follow a long-tailed class distribution without regard to manual balancing.The recognition effect of long-tailed image recognition algorithms based on deep learning is not good, and the recognition accuracy of middle and tail categories is unsatisfactory.In order to address the problems, a Personalized Multi-expert Recognition Algorithm (PMRA) is proposed.Firstly, a multi-expert network is constructed by integrating multiple branches based on the residual network.Then, a personalized learning module is built by assigning personalized training data to different experts to improve the recognition accuracy of middle and tail categories, and a personalized information enhancement module is built by fusing the experts information to deal with the lack of information of middle and tail categories.In the multi-expert network fusing multiple modules, the overall recognition accuracy of long-tailed images is improved by two stages of learning.Finally, the experimental results on benchmark datasets of CIFAR-10-LT, CIFAR-100-LT, ImageNet-LT and iNaturalist2018 show that the recognition accuracy of the proposed algorithm on multiple datasets is better than that of other algorithms.
Get Citation
Copy Citation Text
WU Lei, HAN Hua, HUANG Li, A.A.M.Muzahid. A Personalized Expert Recognition Algorithm for Long-Tailed Images[J]. Electronics Optics & Control, 2023, 30(11): -1
Received: Nov. 26, 2022
Accepted: --
Published Online: Jan. 20, 2024
The Author Email: