Chao Lei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Chao Lei has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Ardhendu Behera
Edge Hill University
2 shared publications - Fuzhong Li
Faculty of Software Technologies, Shanxi Agricultural University, Jinzhong 030801, China
2 shared publications - Reena Reena
Department of Computer Science, Edge Hill University, St Helens Road, Ormskirk, Lancashire L39 4QP, UK
2 shared publications - Wuping Zhang
School of Software, Shanxi Agricultural University, Jinzhong 030801, China
2 shared publications - Xin Lei
Faculty of Software Technology, Shanxi Agricultural University, No. 1 Mingxian South Road, Taiyuan 030801, China
2 shared publications - Yang Sun
Department of Computer Science and Technology, Hebei University, No. 180 Wusi East Road, Baoding 071002, China
2 shared publications
Latest Publications
- Open Access
- 21 August 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (16) , 1790
A Controlled Evaluation of Dual-Channel Feature Enhancement and Multi-Level Knowledge Distillation for Lightweight Plant Disease Recognition
Plant disease symptoms combine local texture changes with patterns distributed across a leaf, while practical recognition models must remain compact. We introduce DC-FEN, a MobileNetV3-based design that models spatial-token relations and channel interactions in parallel and injects them through gated residual fusion. We also examine output-distribution, direct-feature, and token-relation transfer under same-backbone and heterogeneous teachers. PlantVillage and Plant Pathology 2021 (FGVC8) are evaluated with duplicate-audited, group-aware 70/15/15 splits, an explicit unresolved-leaf sensitivity check, validation-only selection, five training seeds, class-sensitive metrics, and paired seed-wise descriptive summaries. On PlantVillage, the no-additional-attention student, DC-FEN teacher, and DC-FEN joint student obtain macro F1 scores of , , and . On FGVC8, the corresponding scores are , , and . At the prespecified FGVC8 threshold of 0.5, DCAB changed sample-wise F1 by percentage points relative to the unmodified backbone; validation-selected global and label-specific thresholds changed this contrast to and points, while threshold-free macro mAP remained essentially unchanged. A duplicate-audited PlantDoc pressure test reduced frozen-checkpoint accuracy to and , showing that external generalization remains unestablished. A ResNet50 teacher gives logit-only students macro F1 on PlantVillage and sample-wise F1 on FGVC8. After separately weighting the direct and relation terms, the corresponding joint students obtain and , recovering the degradation seen with unit internal weights while remaining close to logit-only transfer. Thus, the study evaluates the benefits and limits of explicit spatial–channel interaction and shows that adding intermediate transfer constraints does not guarantee a stronger student.- 21 August 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (16) , 1790
Open AccessA Controlled Evaluation of Dual-Channel Feature Enhancement and Multi-Level Knowledge Distillation for Lightweight Plant Disease Recognition
Plant disease symptoms combine local texture changes with patterns distributed across a leaf, while practical recognition models must remain compact. We introduce DC-FEN, a MobileNetV3-based design that models spatial-token relations and channel interactions in parallel and injects them through gated residual fusion. We also examine output-distribution, direct-feature, and token-relation transfer under same-backbone and heterogeneous teachers. PlantVillage and Plant Pathology 2021 (FGVC8) are evaluated with duplicate-audited, group-aware 70/15/15 splits, an explicit unresolved-leaf sensitivity check, validation-only selection, five training seeds, class-sensitive metrics, and paired seed-wise descriptive summaries. On PlantVillage, the no-additional-attention student, DC-FEN teacher, and DC-FEN joint student obtain macro F1 scores of , , and . On FGVC8, the corresponding scores are , , and . At the prespecified FGVC8 threshold of 0.5, DCAB changed sample-wise F1 by percentage points relative to the unmodified backbone; validation-selected global and label-specific thresholds changed this contrast to and points, while threshold-free macro mAP remained essentially unchanged. A duplicate-audited PlantDoc pressure test reduced frozen-checkpoint accuracy to and , showing that external generalization remains unestablished. A ResNet50 teacher gives logit-only students macro F1 on PlantVillage and sample-wise F1 on FGVC8. After separately weighting the direct and relation terms, the corresponding joint students obtain and , recovering the degradation seen with unit internal weights while remaining close to logit-only transfer. Thus, the study evaluates the benefits and limits of explicit spatial–channel interaction and shows that adding intermediate transfer constraints does not guarantee a stronger student. - Open Access
- 23 June 2026
- Published by MDPI AG in Preprints.org
Dual-Channel Attention and Category-Guided Distillation for Lightweight Plant Disease Recognition
Plant disease recognition remains challenging due to complex imaging conditions, inter-class visual similarity, and the need to deploy accurate models on resource-constrained devices. Although convolutional neural networks (CNNs) are effective at local pattern extraction, their locality bias limits global context modeling. Transformer-based models alleviate this issue, yet many existing designs do not explicitly exploit the complementarity between spatial and channel interactions. To address this limitation, we propose a Dual-Channel Feature Enhancement Network (DC-FEN) as the teacher model, which integrates spatial attention and channel attention to capture lesion-sensitive spatial cues and discriminative channel semantics. To enable lightweight deployment, we further introduce a Category-Guided Knowledge Distillation (CGKD) framework that transfers both category-level semantic knowledge and feature-level relational knowledge from the teacher to a MobileNetV3 student. Experimental results show that the proposed distillation framework consistently improves the performance of the lightweight MobileNetV3 student over its original baseline.
- 23 June 2026
- Published by MDPI AG in Preprints.org
Open AccessDual-Channel Attention and Category-Guided Distillation for Lightweight Plant Disease Recognition
Plant disease recognition remains challenging due to complex imaging conditions, inter-class visual similarity, and the need to deploy accurate models on resource-constrained devices. Although convolutional neural networks (CNNs) are effective at local pattern extraction, their locality bias limits global context modeling. Transformer-based models alleviate this issue, yet many existing designs do not explicitly exploit the complementarity between spatial and channel interactions. To address this limitation, we propose a Dual-Channel Feature Enhancement Network (DC-FEN) as the teacher model, which integrates spatial attention and channel attention to capture lesion-sensitive spatial cues and discriminative channel semantics. To enable lightweight deployment, we further introduce a Category-Guided Knowledge Distillation (CGKD) framework that transfers both category-level semantic knowledge and feature-level relational knowledge from the teacher to a MobileNetV3 student. Experimental results show that the proposed distillation framework consistently improves the performance of the lightweight MobileNetV3 student over its original baseline.