CornViT: A Multi-Stage Convolutional Vision Transformer Framework for Hierarchical Corn Kernel Analysis

2025-12-20

Computers

AI Summary

This study addresses the inefficiency and subjectivity of manual maize kernel grading by proposing a three-stage Convolutional Vision Transformer (CvT) framework that sequentially classifies kernel purity, shape (flat or round), and embryo orientation for pure flat kernels—mimicking expert grading logic. To the best of our knowledge, this is the first application of a multi-stage CvT architecture to fine-grained maize kernel analysis. The authors construct and publicly release three high-quality annotated datasets and leverage an ImageNet-22k pre-trained CvT-13 model, fine-tuning only the classification head for efficient inference. On a custom test set, the model achieves accuracies of 93.76%, 94.11%, and 91.12% on the three respective tasks, significantly outperforming ResNet-50 and DenseNet-121. An interpretable web application is also developed for end-to-end deployment.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Kernel Methods
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate grading of corn kernels is critical for seed certification, directional seeding, and breeding, yet it is still predominantly performed by manual inspection. This work introduces CornViT, a three-stage Convolutional Vision Transformer (CvT) framework that emulates the hierarchical reasoning of human seed analysts for single-kernel evaluation. Three sequential CvT-13 classifiers operate on 384×384 RGB images: Stage 1 distinguishes pure from impure kernels; Stage 2 categorizes pure kernels into flat and round morphologies; and Stage 3 determines the embryo orientation (up vs. down) for pure, flat kernels. Starting from a public corn seed image collection, we manually relabeled and filtered images to construct three stage-specific datasets: 7265 kernels for purity, 3859 pure kernels for morphology, and 1960 pure–flat kernels for embryo orientation, all released as benchmarks. Head-only fine-tuning of ImageNet-22k pretrained CvT-13 backbones yields test accuracies of 93.76% for purity, 94.11% for shape, and 91.12% for embryo-orientation detection. Under identical training conditions, ResNet-50 reaches only 76.56 to 81.02 percent, whereas DenseNet-121 attains 86.56 to 89.38 percent accuracy. These results highlight the advantages of convolution-augmented self-attention for kernel analysis. To facilitate adoption, we deploy CornViT in a Flask-based web application that performs stage-wise inference and exposes interpretable outputs through a browser interface. Together, the CornViT framework, curated datasets, and web application provide a deployable solution for automated corn kernel quality assessment in seed quality workflows. Source code and data are publicly available.

Problem

• corn kernel grading
• seed quality assessment
• automated seed analysis
• hierarchical classification
• embryo orientation detection

Innovation

• Convolutional Vision Transformer
• Hierarchical classification
• Corn kernel analysis
• Embryo orientation detection
• Agricultural AI