RESUMEN

Plant diseases pose significant threats to agriculture, impacting both food safety and public health. Traditional plant disease detection systems are typically limited to recognizing disease categories included in the training dataset, rendering them ineffective against new disease types. Although out-of-distribution (OOD) detection methods have been proposed to address this issue, the impact of fine-tuning paradigms on these methods has been overlooked. This paper focuses on studying the impact of fine-tuning paradigms on the performance of detecting unknown plant diseases. Currently, fine-tuning on visual tasks is mainly divided into visual-based models and visual-language-based models. We first discuss the limitations of large-scale visual language models in this task: textual prompts are difficult to design. To avoid the side effects of textual prompts, we futher explore the effectiveness of purely visual pre-trained models for OOD detection in plant disease tasks. Specifically, we employed five publicly accessible datasets to establish benchmarks for open-set recognition, OOD detection, and few-shot learning in plant disease recognition. Additionally, we comprehensively compared various OOD detection methods, fine-tuning paradigms, and factors affecting OOD detection performance, such as sample quantity. The results show that visual prompt tuning outperforms fully fine-tuning and linear probe tuning in out-of-distribution detection performance, especially in the few-shot scenarios. Notably, the max-logit-based on visual prompt tuning achieves an AUROC score of 94.8 % in the 8-shot setting, which is nearly comparable to the method of fully fine-tuning on the full dataset (95.2 % ), which implies that an appropriate fine-tuning paradigm can directly improve OOD detection performance. Finally, we visualized the prediction distributions of different OOD detection methods and discussed the selection of thresholds. Overall, this work lays the foundation for unknown plant disease recognition, providing strong support for the security and reliability of plant disease recognition systems. We will release our code at https://github.com/JiuqingDong/PDOOD to further advance this field.

Asunto(s)

RESUMEN

Previous work on plant disease detection demonstrated that object detectors generally suffer from degraded training data, and annotations with noise may cause the training task to fail. Well-annotated datasets are therefore crucial to build a robust detector. However, a good label set generally requires much expert knowledge and meticulous work, which is expensive and time-consuming. This paper aims to learn robust feature representations with inaccurate bounding boxes, thereby reducing the model requirements for annotation quality. Specifically, we analyze the distribution of noisy annotations in the real world. A teacher-student learning paradigm is proposed to correct inaccurate bounding boxes. The teacher model is used to rectify the degraded bounding boxes, and the student model extracts more robust feature representations from the corrected bounding boxes. Furthermore, the method can be easily generalized to semi-supervised learning paradigms and auto-labeling techniques. Experimental results show that applying our method to the Faster-RCNN detector achieves a 26% performance improvement on the noisy dataset. Besides, our method achieves approximately 75% of the performance of a fully supervised object detector when 1% of the labels are available. Overall, this work provides a robust solution to real-world location noise. It alleviates the challenges posed by noisy data to precision agriculture, optimizes data labeling technology, and encourages practitioners to further investigate plant disease detection and intelligent agriculture at a lower cost. The code will be released at https://github.com/JiuqingDong/TS_OAMIL-for-Plant-disease-detection.

RESUMEN

Plant disease detection has made significant strides thanks to the emergence of deep learning. However, existing methods have been limited to closed-set and static learning settings, where models are trained using a specific dataset. This confinement restricts the model's adaptability when encountering samples from unseen disease categories. Additionally, there is a challenge of knowledge degradation for these static learning settings, as the acquisition of new knowledge tends to overwrite the old when learning new categories. To overcome these limitations, this study introduces a novel paradigm for plant disease detection called open-world setting. Our approach can infer disease categories that have never been seen during the model training phase and gradually learn these unseen diseases through dynamic knowledge updates in the next training phase. Specifically, we utilize a well-trained unknown-aware region proposal network to generate pseudo-labels for unknown diseases during training and employ a class-agnostic classifier to enhance the recall rate for unknown diseases. Besides, we employ a sample replay strategy to maintain recognition ability for previously learned classes. Extensive experimental evaluation and ablation studies investigate the efficacy of our method in detecting old and unknown classes. Remarkably, our method demonstrates robust generalization ability even in cross-species disease detection experiments. Overall, this open-world and dynamically updated detection method shows promising potential to become the future paradigm for plant disease detection. We discuss open issues including classification and localization, and propose promising approaches to address them. We encourage further research in the community to tackle the crucial challenges in open-world plant disease detection. The code will be released at https://github.com/JiuqingDong/OWPDD.

RESUMEN

BACKGROUND: Scoliosis is a type of spinal deformity, which is harmful to a person's health. In severe cases, it can trigger paralysis or death. The measurement of Cobb angle plays an essential role in assessing the severity of scoliosis. PURPOSE: The aim of this paper is to propose an automatic system for landmark detection and Cobb angle estimation, which can effectively help clinicians diagnose and treat scoliosis. METHODS: A novel hybrid framework was proposed to measure Cobb angle precisely for clinical diagnosis, which was referred as W-Transformer due to its w-shaped architecture. First, a convolutional neural network of cascade residual blocks as our backbone was designed. Then a transformer was fused to learn the dependency information between spine and landmarks. In addition, a reinforcement branch was designed to improve the overlap of landmarks, and an improved prediction module was proposed to fine-tune the final coordinates of landmarks in Cobb angles estimation. Besides, the public Accurate Automated Spinal Curvature Estimation (AASCE) MICCAI 2019 challenge was served as data set. It supplies 609 manually labeled spine anterior-posterior (AP) X-ray images, each of which contains a total of 68 landmark labels and three Cobb Angles tags. RESULTS: From the perspective of the AASCE MICCAI 2019 challenge, we achieved a lower symmetric mean absolute percentage error (SMAPE) of 8.26% for all Cobb angles and the lowest averaged detection error of 50.89 in terms of landmark detection, compared with many state-of-the-art methods. We also provided the SMAPEs for the Cobb angles of the proximal-thoracic (PT), the main-thoracic (MT), and the thoracic-lumbar (TL) area, which are 5.27%, 14.59%, and 20.97% respectively, however, these data were not covered in most previous studies. Statistical analysis demonstrates that our model has obtained a high level of Pearson correlation coefficient of 0.9398 ( p < 0.001 $p<0.001$ ), which shows excellent reliability of our model. Our model can yield 0.9489 ( p < 0.001 $p<0.001$ ), 0.8817 ( p < 0.001 $p<0.001$ ), and 0.9149 ( p < 0.001 $p<0.001$ ) for PT, MT, and TL, respectively. The overall variability of Cobb angle measurement is less than 4 ∘ $^\circ$ , implying clinical value. And the mean absolute deviation (standard deviation) for three regions is 3.64 ∘ $^\circ$ (4.13 ∘ $^\circ$ ), 3.84 ∘ $^\circ$ (4.66 ∘ $^\circ$ ), and 3.80 ∘ $^\circ$ (4.19 ∘ $^\circ$ ). The results of Student paired t $t$ -test indicate that no statistically significant differences are observed between manual measurement and our automatic approach ( p $p$ -value is always > $>$ 0.05). Regarding the diagnosis of scoliosis (Cobb angle > $>$ 10 ∘ $^\circ$ ), the proposed method achieves a high sensitivity of 0.9577 and a specificity of 0.8475 for all spinal regions. CONCLUSIONS: This study offers a brand-new automatic approach that is potentially of great benefit of the complex task of landmark detection and Cobb angle evaluation, which can provide helpful navigation information about the early diagnosis of scoliosis.

Asunto(s)

RESUMEN

Object detection models have become the current tool of choice for plant disease detection in precision agriculture. Most existing research improved the performance by ameliorating networks and optimizing the loss function. However, because of the vast influence of data annotation quality and the cost of annotation, the data-centric part of a project also needs more investigation. We should further consider the relationship between data annotation strategies, annotation quality, and the model's performance. In this paper, a systematic strategy with four annotation strategies for plant disease detection is proposed: local, semi-global, global, and symptom-adaptive annotation. Labels with different annotation strategies will result in distinct models' performance, and their contrasts are remarkable. An interpretability study of the annotation strategy is conducted by using class activation maps. In addition, we define five types of inconsistencies in the annotation process and investigate the severity of the impact of inconsistent labels on model's performance. Finally, we discuss the problem of label inconsistency during data augmentation. Overall, this data-centric quantitative analysis helps us to understand the significance of annotation strategies, which provides practitioners a way to obtain higher performance and reduce annotation costs on plant disease detection. Our work encourages researchers to pay more attention to annotation consistency and the essential issues of annotation strategy. The code will be released at: https://github.com/JiuqingDong/PlantDiseaseDetection_Yolov5 .