Detection of unknown strawberry diseases based on OpenMatch and two-head network for continual learning

Abstract

For continual learning in the process of plant disease recognition it is necessary to first distinguish between unknown diseases from those of known diseases. This paper deals with two different but related deep learning techniques for the detection of unknown plant diseases; Open Set Recognition (OSR) and Out-of-Distribution (OoD) detection. Despite the significant progress in OSR, it is still premature to apply it to fine-grained recognition tasks without outlier exposure that a certain part of OoD data (also called known unknowns) are prepared for training. On the other hand, OoD detection requires intentionally prepared outlier data during training. This paper analyzes two-head network included in OoD detection models, and semi-supervised OpenMatch associated with OSR technology, which explicitly and implicitly assume outlier exposure, respectively. For the experiment, we built an image dataset of eight strawberry diseases. In general, a two-head network and OpenMatch cannot be compared due to different training settings. In our experiment, we changed their training procedures to make them similar for comparison and show that modified training procedures resulted in reasonable performance, including more than 90% accuracy for strawberry disease classification as well as detection of unknown diseases. Accurate detection of unknown diseases is an important prerequisite for continued learning.

Identifying Minerals from Image Using Out-of-Distribution Artificial Intelligence-Based Model

“…In this context, the model distinguishes itself by achieving 67.8% accuracy in classifying out-ofdistribution minerals as unknown categories. This OOD Accuracy is lower than that of other applications listed in references [17][18][19][20][21] because minerals in the same category may have different colors and textures, as shown in Figure 1, while different categories of minerals may also have the same colors and textures [6]. This makes mineral identification more challenging, resulting in similarly lower ID Accuracy than other applications.…”

“…Exemplifying the efficacy of OOD detection, Jiang et al [17] adeptly employed this technique to discern between known and unknown instances of plant diseases. Similarly, Saadati et al [18] conducted OOD detection to support the robustness of insect classification models.…”

Mineral Identification based on Out-Of-Distribution Detection

“…In this context, the model distinguishes itself by achieving a 67.8% accuracy in classifying out-of-distribution minerals as unknown categories. This OOD Accuracy is lower than that of other applications listed in references [17][18][19][20][21] because minerals of the same category may have different colors and textures, while different categories of minerals may have the same colors and textures [6]. This makes mineral identification more challenging, resulting in similarly lower ID Accuracy than other applications.…”

“…Exemplifying the efficacy of OOD detection, Jiang et al [17] adeptly employed this technique to discern between known and unknown instances of plant diseases, while Saadati et al [18] similarly conducted OOD detection to bolster the robustness of insect classification models. Furthermore, the utility of OOD detection extends beyond these domains, showcasing notable promise in the arenas of medical image diagnosis [19], network security [20], and quality control [21].…”

Embracing limited and imperfect training datasets: opportunities and challenges in plant disease recognition using deep learning

“…In general, metric learning pushes models to learn robust feature spaces and thus implicitly contributes to the recognition of the unknown class. In addition, an extra probability branch is explicitly utilized to distinguish between known and unknown classes along with a generic classification branch for known classes ( Jiang et al., 2022 ). Simultaneously, images belonging to unknown classes are utilized to train the models, where exposure to unknown classes is beneficial, although unknown classes in the training stage may also appear in the test stage.…”