Plant diseases present a significant threat to global agricultural productivity, endangering both crop yields and quality. Traditional detection methods largely rely on manual inspection, a process that is not only labor-intensive and time-consuming but also subject to subjective biases and dependent on operators’ expertise. Recent advancements in Transformer-based architectures have shown substantial progress in image classification tasks, particularly excelling in global feature extraction. However, despite their strong performance, the high computational complexity and large parameter requirements of Transformer models limit their practical application in plant disease detection. To address these constraints, this study proposes an optimized Efficient Swin Transformer specifically engineered to reduce computational complexity while enhancing classification accuracy. This model is an improvement over the Swin-T architecture, incorporating two pivotal modules: the Selective Token Generator and the Feature Fusion Aggregator. The Selective Token Generator minimizes the number of tokens processed, significantly increasing computational efficiency and facilitating multi-scale feature extraction. Concurrently, the Feature Fusion Aggregator adaptively integrates static and dynamic features, thereby enhancing the model’s ability to capture complex details within intricate environmental contexts.Empirical evaluations conducted on the PlantDoc dataset demonstrate the model’s superior classification performance, achieving a precision of 80.14% and a recall of 76.27%. Compared to the standard Swin-T model, the Efficient Swin Transformer achieves approximately 20.89% reduction in parameter size while improving precision by 4.29%. This study substantiates the potential of efficient token conversion techniques within Transformer architectures, presenting an effective and accurate solution for plant disease detection in the agricultural sector.

Feature Selection (FS) is a key pre-processing step in pattern recognition and data mining tasks, which can effectively avoid the impact of irrelevant and redundant features on the performance of classification models. In recent years, meta-heuristic algorithms have been widely used in FS problems, so a Hybrid Binary Chaotic Salp Swarm Dung Beetle Optimization (HBCSSDBO) algorithm is proposed in this paper to improve the effect of FS. In this hybrid algorithm, the original continuous optimization algorithm is converted into binary form by the S-type transfer function and applied to the FS problem. By combining the K nearest neighbor (KNN) classifier, the comparative experiments for FS are carried out between the proposed method and four advanced meta-heuristic algorithms on 16 UCI (University of California, Irvine) datasets. Seven evaluation metrics such as average adaptation, average prediction accuracy, and average running time are chosen to judge and compare the algorithms. The selected dataset is also discussed by categorizing it into three dimensions: high, medium, and low dimensions. Experimental results show that the HBCSSDBO feature selection method has the ability to obtain a good subset of features while maintaining high classification accuracy, shows better optimization performance. In addition, the results of statistical tests confirm the significant validity of the method.