In the process of soil image classification, the issue of class imbalance occurs, which leads to a decline in the generalization performance of the classifier due to the lack of data from minority classes. We investigated the effectiveness of Mixup through margin statistical analysis and successfully improved the deep imbalanced classification with uneven margins. Additionally, we investigated the relationship between margins and logits, and empirically discovered that uncalibrated margins exhibit a positive correlation with logits. Based on this revelation, we propose a Mixup-based margin-aware and calibration model to address the challenge of handling imbalanced soil image classification data. We conducted experiments using the Soil dataset and additionally tested the generalization capabilities of our method on the CIFAR10-LT, CIFAR100-LT, and ImageNet-LT datasets. The experimental results indicate that our approach achieved impressive results.

Xin Zheng, Xin Bai. A Mixup-based Margin Aware and Calibration Model for Imbalanced in Soil Classification. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 2: 12-18. https://doi.org/10.25236/AJCIS.2024.070202.

[1] K. Cao, C. Wei, A. Gaidon et al. Learning imbalanced datasets with label-distribution-aware margin loss, Advances in neural information processing systems, vol. 32, 2019.

[2] H. Zhang, M. Cisse, Y. N. Dauphin et al. mixup: Beyond empirical risk minimization, arXiv preprint arXiv:1710.09412, 2017.

[3] Y. Wang, B. Zhang, W. Hou et al. Margin calibration for long-tailed visual recognition, in Asian Conference on Machine Learning, 2023: PMLR, pp. 1101-1116.

[4] B. Kang, S. Xie, M. Rohrbach et al. Decoupling representation and classifier for long-tailed recognition, arXiv preprint arXiv:1910.09217, 2019.

[5] C. Feng, Y. Zhong, and W. Huang Exploring classification equilibrium in long-tailed object detection, IEEE/CVF International conference on computer vision, 2021, pp. 3417-3426.

[6] Y. Zang, C. Huang, and C. C. Loy Fasa: Feature augmentation and sampling adaptation for long-tailed instance segmentation, IEEE/CVF International Conference on Computer Vision, 2021, pp. 3457-3466.

[7] Y. Zhang, B. Kang, B. Hooi et al. Deep long-tailed learning: A survey, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.

[8] Y. Cui, M. Jia, T.-Y. Lin et al. Class-balanced loss based on effective number of samples, IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 9268-9277.

[9] J. Ren, C. Yu, X. Ma et al. Balanced meta-softmax for long-tailed visual recognition, Advances in neural information processing systems, vol. 33, pp. 4175-4186, 2020.

[10] T.-Y. Lin, P. Goyal, R. Girshick et al. Focal loss for dense object detection, IEEE international conference on computer vision, 2017, pp. 2980-2988.

[11] D. Cao, X. Zhu, X. Huang et al. Domain balancing: Face recognition on long-tailed domains, IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 5671-5679.

[12] T. Wu, Z. Liu, Q. Huang et al. Adversarial robustness under long-tailed distribution, IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 8659-8668.

[13] J. Tan, C. Wang, B. Li et al. Equalization loss for long-tailed object recognition, IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 11662-11671.

[14] B. Zhou, Q. Cui, X.-S. Wei et al. Bbn: Bilateral-branch network with cumulative learning for long-tailed visual recognition, IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 9719-9728.

[15] H.-P. Chou, S.-C. Chang, J.-Y. Pan et al. Remix: rebalanced mixup, in Computer Vision–ECCV 2020 Workshops: Glasgow, UK, August 23–28, 2020, Proceedings, Part VI 16, 2020: Springer, pp. 95-110.

[16] Y. Hong, S. Han, K. Choi et al. Disentangling label distribution for long-tailed visual recognition, IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 6626-6636.

[17] S. Zhang, Z. Li, S. Yan et al. Distribution alignment: A unified framework for long-tail visual recognition, IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 2361-2370.

[18] P. Wang, K. Han, X.-S. Wei et al. Contrastive learning based hybrid networks for long-tailed image classification, IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 943-952.

[19] X. Chen, Y. Zhou, D. Wu et al. Imagine by reasoning: A reasoning-based implicit semantic data augmentation for long-tailed classification, AAAI Conference on Artificial Intelligence, 2022, vol. 36, no. 1, pp. 356-364.

[20] S. Zhang, C. Chen, X. Zhang et al. Label-occurrence-balanced mixup for long-tailed recognition, in ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022: IEEE, pp. 3224-3228.

[21] J. Wang, W. Zhang, Y. Zang et al. Seesaw loss for long-tailed instance segmentation, IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 9695-9704.