{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T06:45:43Z","timestamp":1782456343503,"version":"3.54.5"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T00:00:00Z","timestamp":1782432000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Plant Sci."],"abstract":"<jats:p>Accurate classification of unsound maize kernels is critical for grain quality detection and grading. Single red-green-blue (RGB) images are incapable of effectively capturing variations in the internal chemical components of maize kernels, which easily causes category confusion. Meanwhile, hyperspectral images (HSI) lacks the spatial structural information and exhibits the limited discriminative ability when used independently. To address these challenges, this paper proposes a multimodal hybrid enhancement network (MHENet) with a parallel spectral-spatial dual-branch architecture. The proposed network adopts an optimized 1DCNN-3 module for hyperspectral sequence feature extraction and MobileNetV4-Small for lightweight RGB image spatial feature extraction, enabling the comprehensive capture of both spectral physicochemical characteristics and external morphological appearance features of maize kernels. Specifically, the cross-modal projection embedding (CMPE) module achieves heterogeneous feature space alignment; the bidirectional Mamba attention (BMA) module extracts global contextual information with linear complexity; and the entropy-driven adaptive gating (EAG) fusion module enables dynamic weighted fusion of multimodal features. Experiments are conducted on a self-constructed RGB and HSI dataset of unsound maize kernels. The experimental results demonstrate that the proposed MHENet achieves an accuracy of 97.68%, a macro-P of 97.61%, a macro-R of 97.28%, and a macro-F1 of 97.44% in six-category maize kernel classification. Five-fold cross-validation results in an average accuracy of 97.62% and a standard deviation of 0.14. Compared with feature concatenation, Hadamard product, and cross-attention fusion strategies, the proposed MHENet outperforms these methods in terms of accuracy, lightweight, and stability. This work provides an essential guidance to accurately classify unsound maize kernels.<\/jats:p>","DOI":"10.3389\/fpls.2026.1867535","type":"journal-article","created":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T05:45:53Z","timestamp":1782452753000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["MHENet: a multimodal hybrid enhancement network for accurate classification of unsound maize kernels"],"prefix":"10.3389","volume":"17","author":[{"given":"Chenxia","family":"Wan","sequence":"first","affiliation":[{"name":"Key Laboratory of Grain Information Processing and Control (Henan University of Technology), Ministry of Education","place":["Zhengzhou, China"]},{"name":"Henan Key Laboratory of Grain Storage Information Intelligent Perception and Decision Making, Henan University of Technology","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruitao","family":"Li","sequence":"additional","affiliation":[{"name":"Key Laboratory of Grain Information Processing and Control (Henan University of Technology), Ministry of Education","place":["Zhengzhou, China"]},{"name":"Henan Key Laboratory of Grain Storage Information Intelligent Perception and Decision Making, Henan University of Technology","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenzheng","family":"Li","sequence":"additional","affiliation":[{"name":"Key Laboratory of Grain Information Processing and Control (Henan University of Technology), Ministry of Education","place":["Zhengzhou, China"]},{"name":"Henan Key Laboratory of Grain Storage Information Intelligent Perception and Decision Making, Henan University of Technology","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peng","family":"Li","sequence":"additional","affiliation":[{"name":"Institute for Complexity Science, Henan University of Technology","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,26]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1080\/14735903.2025.2569166","article-title":"Optical imaging for food grain quality evaluation - recent advances and future perspectives","volume":"23","author":"Ageh","year":"2025","journal-title":"Int. J. Agric. Sustainability"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.3390\/agronomy12081843","article-title":"Development of deep learning methodology for maize seed variety recognition based on improved swin transformer","volume":"12","author":"Bi","year":"2022","journal-title":"Agronomy-Basel"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2026.111675","article-title":"Kernel loss prediction and parameter optimization for seed-maize husking devices using MFBD-DEM and bi-level BO-ISSA-BP","volume":"246","author":"Chang","year":"2026","journal-title":"Comput. Electron. Agric."},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.foodchem.2026.147961","article-title":"Advances and challenges in multidimensional architectural applications of 1D\/2D\/3D convolutional neural networks in food quality assessment","volume":"505","author":"Deng","year":"2026","journal-title":"Food Chem."},{"key":"B5","doi-asserted-by":"publisher","first-page":"2971","DOI":"10.1007\/s11694-025-03888-1","article-title":"The multilayer perceptron recognition model for imperfect maize kernels based on hyperspectral imaging technology","volume":"20","author":"Dong","year":"2026","journal-title":"J. Food Meas. Charact."},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.3390\/foods13244044","article-title":"Maize kernel broken rate prediction using machine vision and machine learning algorithms","volume":"13","author":"Fan","year":"2024","journal-title":"Foods"},{"key":"B7","doi-asserted-by":"publisher","first-page":"126722","DOI":"10.2139\/ssrn.5245437","article-title":"Efficient wheat variety identification using raman hyperspectral imaging in combination with deep learning","volume":"344","author":"Fan","year":"2026","journal-title":"Spectrochimica Acta Part. A-Molecular Biomolecular Spectrosc."},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture14101780","article-title":"Detection of mechanical damage in corn seeds using hyperspectral imaging and the ResNeSt_E deep learning network","volume":"14","author":"Huang","year":"2024","journal-title":"Agriculture-Basel"},{"key":"B9","doi-asserted-by":"publisher","first-page":"16076","DOI":"10.1109\/TNNLS.2023.3291793","article-title":"Flexible label-induced manifold broad learning system for multiclass recognition","volume":"35","author":"Jin","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2025.111321","article-title":"Deep learning-driven hyperspectral imaging for drought stress detection in dragoon lettuce for space production","volume":"242","author":"Kim","year":"2026","journal-title":"Comput. Electron. Agric."},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.3390\/info15070408","article-title":"Application of attention-enhanced 1D-CNN algorithm in hyperspectral image and spectral fusion detection of moisture content in Orah mandarin (Citrus reticulata Blanco)","volume":"15","author":"Li","year":"2024","journal-title":"Information"},{"key":"B12","doi-asserted-by":"publisher","first-page":"111565","DOI":"10.1016\/j.compag.2026.111565","article-title":"Physics-guided spatiotemporal graph transformer for explainable detection and risk forecasting of mold and condensation in grain storage","volume":"245","author":"Li","year":"2026","journal-title":"Comput. and Electron. Agric."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.25165\/j.ijabe.20261901.10322","article-title":"Ai-driven technologies for pest monitoring, unsound kernel detection, and intelligent aeration in grain storage","volume":"19","author":"Luo","year":"2026","journal-title":"Int. J. Agric. Biol. Eng."},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1016\/j.foodcont.2022.108819","article-title":"Determination of wheat kernels damaged by fusarium head blight using monochromatic images of effective wavelengths from hyperspectral imaging coupled with an architecture self-search deep network","volume":"135","author":"Lv","year":"2022","journal-title":"Food Control"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1016\/j.jfca.2025.108018","article-title":"Detection of maize seed viability using time series multispectral imaging technology","volume":"147","author":"Meng","year":"2025","journal-title":"J. Food Compos. Anal."},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1016\/j.foodres.2025.117983","article-title":"A quantitative detection method for maize kernel broken rate based on the optimisation of the MSA transformer algorithm","volume":"224","author":"Qiao","year":"2026","journal-title":"Food Res. Int."},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1016\/j.jfca.2026.109078","article-title":"Nondestructive detection of ganoderic acids based on short-wave infrared hyperspectral and machine learning","volume":"153","author":"Ran","year":"2026","journal-title":"J. Food Compos. Anal."},{"key":"B18","doi-asserted-by":"publisher","first-page":"194","DOI":"10.25165\/j.ijabe.20201305.4413","article-title":"Determination of damaged wheat kernels with hyperspectral imaging analysis","volume":"13","author":"Shao","year":"2020","journal-title":"Int. J. Agric. Biol. Eng."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.3390\/foods15020404","article-title":"Corn kernel segmentation and damage detection using a hybrid watershed-convex hull approach","volume":"15","author":"Shen","year":"2026","journal-title":"Foods"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1016\/j.foodres.2025.115810","article-title":"Discrimination of unsound soybeans using hyperspectral imaging: A deep learning method based on dual-channel feature fusion strategy and attention mechanism","volume":"203","author":"Sun","year":"2025","journal-title":"Food Res. Int."},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2025.109926","article-title":"Dynamic real-time detection for corn kernel breakage rate based on deep learning and sliding window technology","volume":"232","author":"Wang","year":"2025","journal-title":"Comput. Electron. Agric."},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2022.730190","article-title":"Corn seed defect detection based on watershed algorithm and two-pathway convolutional neural networks","volume":"13","author":"Wang","year":"2022","journal-title":"Front. Plant Sci."},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/j.foodchem.2025.146910","article-title":"Food defect detection technologies based on deep learning and prospects in detection of unsound wheat kernels","volume":"496","author":"Wang","year":"2025","journal-title":"Food Chem."},{"key":"B24","doi-asserted-by":"publisher","first-page":"e70424","DOI":"10.1111\/jfpe.70424","article-title":"Fresh corn appearance detection using Vis-NIR fusion and FCD-GELAN","volume":"49","author":"Xu","year":"2026","journal-title":"J. Food Process Eng."},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture14040618","article-title":"Maize kernel quality detection based on improved lightweight YOLOv7","volume":"14","author":"Yang","year":"2024","journal-title":"Agriculture-Basel"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1016\/j.saa.2024.124166","article-title":"Non-destructive detection of defective maize kernels using hyperspectral imaging and convolutional neural network with attention module","volume":"313","author":"Yang","year":"2024","journal-title":"Spectrochimica Acta Part. A-Molecular Biomolecular Spectrosc."},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.3390\/pr13020346","article-title":"Research on mass prediction of maize kernel based on machine vision and machine learning algorithm","volume":"13","author":"Yu","year":"2025","journal-title":"Processes"},{"key":"B28","doi-asserted-by":"publisher","first-page":"109587","DOI":"10.1016\/j.compag.2024.109587","article-title":"Visual large language model for wheat disease diagnosis in the wild","volume":"227","author":"Zhang","year":"2024","journal-title":"Comput. and Electron. Agric."},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2025.111716","article-title":"A novel lightweight model integrating convolutional neural network and self-attention mechanism for corn seeds quality image recognition","volume":"159","author":"Zhang","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"B30","doi-asserted-by":"publisher","first-page":"5833","DOI":"10.3233\/JIFS-213195","article-title":"Unsound wheat kernel recognition based on deep convolutional neural network transfer learning and feature fusion","volume":"43","author":"Zhang","year":"2022","journal-title":"J. Intell. Fuzzy Syst."},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1016\/j.jspr.2025.102722","article-title":"CNN-transformer fusion for improving maize kernel storage quality: A new approach to reducing storage risks","volume":"114","author":"Zhao","year":"2025","journal-title":"J. Stored Prod. Res."},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture15010079","article-title":"Real-time corn variety recognition using an efficient DenXt architecture with lightweight optimizations","volume":"15","author":"Zhao","year":"2025","journal-title":"Agriculture-Basel"},{"key":"B33","doi-asserted-by":"publisher","first-page":"565","DOI":"10.1016\/j.aiia.2025.11.010","article-title":"Sclformer: A synergistic convolution-linear attention transformer for hyperspectral image classification of mechanical damage in maize kernels","volume":"16","author":"Zheng","year":"2026","journal-title":"Artif. Intell. Agric."},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.3390\/foods13101570","article-title":"Discrimination of new and aged seeds based on on-line near-infrared spectroscopy technology combined with machine learning","volume":"13","author":"Zhu","year":"2024","journal-title":"Foods"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1111\/jfpe.14499","article-title":"Detection of corn unsound kernels based on GAN sample enhancement and improved lightweight network","volume":"47","author":"Zhu","year":"2024","journal-title":"J. Food Process Eng."}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1867535\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T05:45:54Z","timestamp":1782452754000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1867535\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,26]]},"references-count":35,"alternative-id":["10.3389\/fpls.2026.1867535"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1867535","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,26]]},"article-number":"1867535"}}