Few-shot cotton leaf spots disease classification based on metric learning

Abstract

Background Cotton diceases seriously affect the yield and quality of cotton. The type of pest or disease suffered by cotton can be determined by the disease spots on the cotton leaves. This paper presents a few-shot learning framework that can be used for cotton leaf disease spot classification task. This can be used in preventing and controlling cotton diseases timely. First, disease spots on cotton leaf’s disease images are segmented by different methods, compared by using support vector machine (SVM) method and threshold segmentation, and discussed the suitable one. Then, with segmented disease spot images as input, a disease spot dataset is established, and the cotton leaf disease spots were classified using a classical convolutional neural network classifier, the structure and framework of convolutional neural network had been designed. At last, the features of two different images are extracted by a parallel two-way convolutional neural network with weight sharing. Then, the network uses a loss function to learn the metric space, in which similar leaf samples are close to each other and different leaf samples are far away from each other. In summary, this work can be regarded as a significang reference and the benchmark comparison for the follow-up studies of few-shot learning tasks in the agricultural field. Results To achieve the classification of cotton leaf spots by small sample learning, a metric-based learning method was developed to extract cotton leaf spot features and classify the sick leaves. The threshold segmentation and SVM were compared in the extracting of leaf spot. The results showed that both of these two method can extract the leaf spot in a good performance, SVM expented more time, but the leaf spot which extracted from SVM was much more suitable for classifying, thus SVM method can retain much more information of leaf spot, such as color, shape, textures, ect, which can help classficating the leaf spot. In the process of leaf spot classification, the two-way parallel convolutional neural network was established for building the leaf spot feature extractor, and feature classifier is constructed. After establishing the metric space, KNN was used as the spot classifier, and for the construction of convolutional neural networks, commonly used models were selected for comparison, and a spatial structure optimizer (SSO) is introduced for local optimization of the model, include Vgg, DesenNet, and ResNet. Experimentally, it is demonstrated that the classification accuracy of DenseNet is the highest, compared to the other two networks, and the classification accuracy of S-DenseNet is 7.7% higher then DenseNet on average for different number of steps. Conclusions As the step increasing, the accuracy of DesenNet, and ResNet are all improved, and after using SSO, each of these neural networks can achieved better performance. But The extent of increase varies, DesenNet with SSO had been improved the most obviously.

Artificial-intelligence and sensing techniques for the management of insect pests and diseases in cotton: a systematic literature review

“…The results were better for the processing of images compared with traditional approaches such as SVM, KNN, ANN and neuro-fuzzy. Liang (2021) also used CNNs (Vgg, DesenNet, ResNet and S-DesneNet). These CNNs were optimized with the spatial structure optimizer (SSO).…”

A survey of few-shot learning in smart agriculture: developments, applications, and challenges

“…At present, few-shot learning is also the most widely used in agricultural and plant properties for the identification of plant diseases. For example, Liang et al Used the few-shot learning method based on metric learning to identify cotton leaf spots [ 19 ], Wang et al proposed multi-mode collaborative representation learning based on disease images and disease texts to solve the problem of vegetable disease identification under complex background [ 52 ], Argüeso et al also used the few-shot model based on metric learning to identify 38 plant diseases in the dataset PlantVillage [ 53 ], and Zhong et al used the conditional adversary automatic encoder (CAAE) to identify citrus golden grape diseases [ 54 ]. These studies only use a small number of labeled samples to achieve satisfactory results, so that the identification of plant diseases will no longer rely on expert experience and realize automatic identification in the future.…”

“…Now there have been many studies on agricultural few-shot learning, whose models can successfully identify crop pests [ 18 ], plant disease [ 19 ], plant breeding [ 20 ] and so on. The emergence of few-shot learning has successfully brought AI into the era of few-shot.…”

Distance-Entropy: An Effective Indicator for Selecting Informative Data

“…However, more commonly used data sources in artificial intelligence (AI)-driven applications are images or videos. For example, based on the RGB image processing or hyperspectral image processing, there have been numerous typical studies and applications, including agricultural yield forecasting (Khaki and Wang, 2019 ; Shahhosseini et al, 2020 ; Jarlan et al, 2021 ), crop pests and diseases identification (Li and Chao, 2020 ; Li et al, 2020 ; Liu and Wang, 2020 ; Liang, 2021 ), agricultural robot and navigation (Wen et al, 2020 ; Zhang et al, 2020 ; Emmi et al, 2021 ), counting of plant fruits (Lin and Guo, 2020 ; Fu et al, 2021 ), etc.…”