Lightweight vision transformer and ResNet-9 models for real-time plant disease detection and pest classification with SHAP explainability
Published in Bioengineering & Biotechnology, Earth & Environment, and Sustainability
Context
The rapid advancement of digital technologies alongside increased global commitment to sustainability has intensified the need for efficient crop management solutions. Across agricultural systems, delayed detection and misclassification of plant diseases remain major contributors to reduced yields, threatening food security and undermining progress toward Sustainable Development Goals such as Zero Hunger, No Poverty, Good Health and Well-being, Climate Action, and Life on Land. Plant pests, diseases, and excessive chemical use further exacerbate these challenges. Early, automated visual detection offers a pathway to environmentally responsible and economically viable agricultural practices.
Objective
This study aims to develop and evaluate an automated, real-time plant disease classification framework using Vision Transformers (ViT) and hybrid ViT–CNN architectures, with the goal of supporting farmers and agronomists in early decision-making and sustainable crop protection.
Methods
The research employs deep learning techniques, including ViT and a combined ViT–CNN model built on ResNet-9, trained and evaluated using four publicly available datasets: the Turkey Plant Pests and Diseases (TPPD) dataset (15 classes), the Namibia Maize Image Dataset (3 classes), the Banana Image Dataset (3 classes), and the Tanzania Maize Dataset (3 classes). SHapley Additive exPlanations (SHAP) were applied to generate saliency maps for interpretability. Comparative analyses assessed performance, accuracy, and classification speed across attention-based and hybrid architectures.
Results and conclusions
The proposed model achieved strong performance with 97.4% accuracy, 96.4% precision, 97.09% recall, a 95.7% F1-score, and high agreement measured by Cohen’s Kappa, outperforming existing benchmark models. SHAP visualizations highlighted that the model leverages high-activation areas, edge features, color patterns, texture, shape, and contextual cues in its predictions. While attention-based models improved accuracy, they also caused reduced classification speed. However, integrating attention blocks with CNN layers effectively compensated for this slowdown, achieving both high accuracy and efficient inference. To evaluate the interpretability and deployment feasibility of the proposed model, we considered several parameters are considered to have an integration of faithfulness, localization quality, sparsity, latency, and energy of the models, including pointing accuracy, localization IoU, Centroid Localization Error (pixels), Attribution Sparsity (%), Insertion AUC, Deletion AUC, Time per Explanation (ms/image), Energy Consumption (J/image), Memory Footprint (MB).
Significance
This research provides a transparent and high-performing deep learning solution for plant disease classification, promoting sustainable agricultural development. By reducing reliance on excessive pesticide and herbicide use, enhancing early diagnosis, and improving decision-making, the model supports environmentally responsible farming practices and contributes to global efforts toward food security and ecological resilience. The significance of this evaluation is that it comprehensively assesses the model’s interpretability and real-world deployability by measuring explanation reliability (faithfulness), spatial precision (localization quality), efficiency (latency and memory), and sustainability (energy consumption), ensuring the model is not only accurate but also transparent, efficient, and practical for deployment. Details at: https://link.springer.com/article/10.1186/s12870-026-08667-8
Wasswa Shafik, born in Makindye, Kampala City, Uganda, is a Computer Scientist, Information Technologist, Founder, and Research Director at the Dig Connectivity Research Laboratory (DCRLab), Kampala, Uganda. He received his Bachelor’s in information technology engineering with a minor in mathematics at Ndejje University, Luweero, Uganda, and a Master’s in information technology engineering (communication and computer networks option) at Yazd University, Yazd, Islamic Republic of Iran. He pursued his PhD in digital science (computer science) at the School of Digital Science, Universiti Brunei Darussalam, Brunei Darussalam. His research broadly examines, integrates, and focuses on developing computationally and statistically efficient models and algorithms to address complex questions about artificial intelligence and machine learning problems. His specific research interests include sustainable development, smart agriculture, computer vision, ecological informatics, applied artificial intelligence, the Internet of Things, cybersecurity and privacy, and smart computing. He has authored, coauthored, edited, coedited, and published hundreds of peer-reviewed books, technical papers, and journal papers, and numerous IEEE international conference papers. His “Hirsch index” is 20+ (source: Scopus Author ID 57210203704). He is the coeditor of Blockchain Technologies for Smart Circular Economy and Organisational Sustainability and Transforming Healthcare Sector Through Artificial Intelligence and Environmental Sustainability (Springer Nature, 2025 and 2024, respectively). He is a member of IEEE and served as a reviewer of several international journals indexed in Scopus, Compendex (Elsevier Engineering Index), and WoS He served in different capacities as department support for mathematics for data science, physical computing for intelligence systems, advanced topics in computing, advanced algorithms, and system performance and evaluation. Prior to this, as a department fellow, he served as a research associate at the Intelligent Network Laboratory in Iran. He served in different capacities as a Community Data Officer at the Programme for Accessible Health, Communication and Education; Research Associate and Data Manager at Population Services International; Data Manager and Research Assistant at the Socio-economic Data Center; Research Lead at TechnoServe; and Ag. Chief Executive Officer at Asmaah Charity Organisation.
Follow the Topic
-
BMC Plant Biology
This is an open access, peer-reviewed journal that considers articles on all aspects of plant biology, including molecular, cellular, tissue, organ and whole organism research.
What are SDG Topics?
An introduction to Sustainable Development Goals (SDGs) Topics and their role in highlighting sustainable development research.
Continue reading announcementRelated Collections
With Collections, you can get published faster and increase your visibility.
Plant disease resistance - Part II
In nature, plants are constantly exposed to a multitude of microorganisms and pathogens, causing diseases that represent a major threat to agriculture, global food security and ecological biodiversity. To combat such diseases, plants have evolved a sophisticated innate immunity system capable of detecting and responding to pathogen infection, as well as acquired immune mechanisms that enable them to recognize and combat pathogens. Novel approaches and recent technological advances such as high‑throughput sequencing and (multi-)omics, single‑cell profiling, CRISPR/Cas9 genome editing, deep learning and precision breeding, among others, can help researchers investigating plant-pathogen interactions and plant diseases with unprecedented resolution and scale.
In support of United Nations' Sustainable Development Goal 2 (SDG 2: Zero Hunger), 13 (SDG 13: Climate action) and 15 (SDG 15: Life on Land), BMC Plant Biology presents the Collection Plant disease resistance - Part II. This Collection invites research that can improve our understanding of disease resistance in plants and plant-pathogen interactions and builds towards the sustainable breeding for disease-resilient crops. Research without a clear focus on plants and crops or plant-microbe interactions will not be considered. We invite submission of research articles that explore, but are not limited to, the following topics:
- Mechanisms and evolution of plant disease resistance: regulation and evolution of resistance genes (R genes), their mode of action, and how R genes initiate immune signaling
- Plant immune systems, immune mechanisms, signaling, systemic acquired resistance, immune priming and immune memory
- Immune surveillance systems in crops
- Biological control agents and induced systemic resistance (ISR)
- Cell-surface immune receptors (e.g. receptor-like kinases (RKs or RLKs) and receptor-like proteins (RPs or RLPs)), and intracellular immune receptors (e.g. the nucleotide-binding domain and leucine-rich repeat-containing protein (NLR) family)
- Mechanisms and regulation of plant hypersensitive response (HR)
- Resistance genes and elicitors of plant immunity
- Plant resistomes (genome-wide repertoires and diversity of plant immune receptors and resistance-associated genes)
- Cell and molecular biology of plant disease resistance
- Plant-pathogen interactions
- Emerging and re-emerging plant pathogens and resistance strategies
- Plant microbiome and endophyte-mediated disease resistance
- The role of rhizosphere microbiome in disease suppression
- Plant secondary metabolites and natural products in disease resistance
- Epigenetic regulation of plant disease resistance
- Small RNAs, RNA silencing and cross-kingdom RNA communication in plant immunity
- RNA-based disease management including RNAi, spray-induced gene silencing (SIGS), host-induced gene silencing (HIGS)
- Cross-kingdom RNA trafficking and RNA-mediated plant-pathogen communication
- Pattern-triggered immunity (PTI) and effector-triggered immunity (ETI)
- Multi-omics approaches, single-cell and spatial transcriptomics and gene-editing for understanding disease resistance
- Spatial, single-cell and cell-type-specific approaches to plant immunity
- Extracellular vesicles and intercellular communication during plant–pathogen interactions
- The influence of climate change and combined abiotic–biotic stresses on disease resistance
- Novel breeding strategies for plant disease resistance
- Bioengineering, genetic modifications and synthetic biology applied to crops to improve disease resistance
- Nanotechnology in plant disease resistance (use of immunity booster nanoparticle formulations)
- Artificial intelligence in plant disease resistance, including plant-microbe interactions
- Technological and methodological advances applied to plant disease resistance research
All manuscripts submitted to this journal, including those submitted to collections and special issues, are assessed in line with our editorial policies and the journal’s peer-review process. Reviewers and editors are required to declare competing interests and can be excluded from the peer review process if a competing interest exists.
Publishing Model: Open Access
Deadline: May 18, 2027
Plant carbon metabolism
Carbon metabolism is central to the growth, development and productivity of photosynthetic organisms. It involves dynamic regulation and interplay between a number of complex processes (e.g. photosynthesis, carbon use and storage, respiration, nutrient metabolism), response and adaptation to environmental cues and stress (e.g. light, CO2 content, temperature), as well as the integration of fine-tuned enzymatic activity, post-translational modifications and signalling pathways.
Over the past decades, much research has been carried out to understand carbon fixation and utilization in model plants/crops, as well as other photosynthetic organisms. Advancing our understanding of the working principles and regulation of key enzymes like Rubisco, the Calvin–Benson–Bassham (CBB) cycle, C3, C4, and CAM metabolism, carbon concentrating mechanisms (CCMs), or how specific micro-compartments and organelles facilitate carbon concentration, has fundamental relevance. Moreover, the engineering of carbon metabolism has the potential to help addressing current global challenges linked to food security and climate change. For instance, novel insights into carbon metabolism regulation can inform breeding strategies to enhance crop yield and resilience by optimizing photosynthetic efficiency and carbon utilization, and lead to metabolic engineering of photosynthetic (micro)organisms with improved carbon fixation.
In support of the United Nations' Sustainable Development Goal 2 (SDG 2: Zero Hunger), BMC Plant Biology launches the Collection Plant carbon metabolism. This Collection invites research addressing the mechanisms and regulation of plant carbon metabolism, including research that explores new applications and approaches to modulate carbon metabolism and enhance the yield and stress resilience in crops and photosynthetic organisms (e.g. micro/macroalgae, cyanobacteria). We invite researchers and experts in the field to submit research articles that explore, but are not limited to, the following topics:
- C3, C4 and CAM metabolism in photosynthetic organisms: diversity, regulation and evolution
- Mechanisms and regulation of the Calvin-Benson-Bassham (CBB) cycle in photosynthetic organisms
- Synthesis, assembly, mechanisms and activity regulation of Rubisco
- Functions and diversity of carbon-concentrating mechanisms (CCMs) in photosynthetic organisms
- Functions, working principles and diversity of micro-compartments and organelles that facilitate carbon concentration in photosynthetic organisms (e.g. carboxysome, pyrenoid)
- Mechanisms and regulation of carbon allocation (e.g. starch and sucrose metabolism)
- Mechanisms and regulation of the central carbon metabolism, e.g. glycolysis, Oxidative Pentose Phosphate pathway, the Tricarboxylic Acid (TCA) cycle in photosynthetic organisms
- Metabolic engineering of carbon metabolism to enhance yield and stress resilience in photosynthetic organisms
- Synthetic biology approaches to enhance carbon fixation
- Impact of environmental stress on carbon metabolism
- Regulation of plant carbon assimilation metabolism by post-translational modifications
- Role of plant secondary metabolites in carbon metabolism regulation
All manuscripts submitted to this journal, including those submitted to collections and special issues, are assessed in line with our editorial policies and the journal’s peer-review process. Reviewers and editors are required to declare competing interests and can be excluded from the peer review process if a competing interest exists.
Publishing Model: Open Access
Deadline: Dec 04, 2026