Improving the Recognition of Known and Unknown Plant Disease Classes Using Deep Learning
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Recently, there has been a growing emphasis on identifying both known and unknown diseases in plant disease recognition. In this task, a model trained only on images of known classes is required to classify an input image into either one of the known classes or into an unknown class. Consequently, the capability to recognize unknown diseases is critical for model deployment. To enhance this capability, we are considering three factors. Firstly, we propose a new logits-based scoring function for unknown scores. Secondly, initial experiments indicate that a compact feature space is crucial for the effectiveness of logits-based methods, leading us to employ the AM-Softmax loss instead of Cross-entropy loss during training. Thirdly, drawing inspiration from the efficacy of transfer learning, we utilize a large plant-relevant dataset, PlantCLEF2022, for pre-training a model. The experimental results suggest that our method outperforms current algorithms. Specifically, our method achieved a performance of 97.90 CSA, 91.77 AUROC, and 90.63 OSCR with the ResNet50 model and a performance of 98.28 CSA, 92.05 AUROC, and 91.12 OSCR with the ConvNext base model. We believe that our study will contribute to the community.
Abstract
Recently, there has been a growing emphasis on identifying both known and unknown diseases in plant disease recognition. In this task, a model trained only on images of known classes is required to classify an input image into either one of the known classes or into an unknown class. Consequently, the capability to recognize unknown diseases is critical for model deployment. To enhance this capability, we are considering three factors. Firstly, we propose a new logits-based scoring function for unknown scores. Secondly, initial experiments indicate that a compact feature space is crucial for the effectiveness of logits-based methods, leading us to employ the AM-Softmax loss instead of Cross-entropy loss during training. Thirdly, drawing inspiration from the efficacy of transfer learning, we utilize a large plant-relevant dataset, PlantCLEF2022, for pre-training a model. The experimental results suggest that our method outperforms current algorithms. Specifically, our method achieved a performance of 97.90 CSA, 91.77 AUROC, and 90.63 OSCR with the ResNet50 model and a performance of 98.28 CSA, 92.05 AUROC, and 91.12 OSCR with the ConvNext base model. We believe that our study will contribute to the community.
참고문헌 (18)
- Savary, S., "The global burden of pathogens and pests on major food crops", Nat. Ecol. E, 3(3), pp.430~439 (2019)
- Thakur, P. S., "Trends in vision-based machine learning techniques for plant disease identification: A systematic review", Expert Syst, 208 (2022)
- Abade, André, "Plant diseases recognition on images using convolutional neural networks : A systematic review", Computers and Electronics in Agriculture, 185 (2021)
- Xu, M., "Transfer learning for versatile plant disease recognition with limited data", Front. Plant Sci, 13 (2022)
- Dong J, "Towards Improved Performance on Plant Disease Recognition with Symptoms Specific Annotation", Smart Media Journal, 11(4), pp.38~45 (2022)
- Meng, Yao, "Known and unknown class recognition on plant species and diseases", Computers and Electronics in Agriculture, 215 (2023)
- "Enhanced Plant Disease Recognition with Limited Training Dataset Using Image Translation and Two-Step Transfer Learning", Ph.D.
- Xu, M., "Embracing limited and imperfect training datasets: opportunities and challenges in plant disease recognition using deep learning", Front. Plant Sci, 14 (2023)
- Xu, M., "A comprehensive survey of image augmentation techniques for deep learning", Pattern Recognit, 137 (2023)
- Geng, C., "Recent advances in open set recognition: A survey", IEEE Trans. Pattern Anal. Mach. Intell, 43(10), pp.3614~3631 (2021)
- Dong J, "Novel Category Discovery in Plant Species and Disease Identification through Knowledge Distillation", Smart Media Journal, 13(7), pp.36~44
- Wang,F., "Additive Margin Softmax for Face Verification", IEEE Signal Process. Lett, 25, pp.926~930 (2018)
- Xu, Mingle, "Unsupervised transfer learning for plant anomaly recognition", Smart Media Journal, 11(4), pp.30~37 (2022)
- Goëau, H., "Overview of PlantCLEF 2022: “Image-based plant identification at global scale", Proceedings of the CLEF 2022-Conference and Labs of the Evaluation Forum, 3180, pp.1916~1928 (2022)
- Goëau, H., "Image-based plant identification at global scale", Proceedings of the CLEF 2022-Conference and Labs of the Evaluation Forum, 3180, pp.1916~1928 (2022)
- Neal, L., "Open set learning with counterfactual images", Proceedings of the European Conference on Computer Vision (ECCV), pp.613~628 (2018)
- Dhamija, A. R., "Reducing network agnostophobia", Adv. Neural Inf. Process. Syst, 31 (2018)