Wheat Variety Classification Using Spatial Deep Convolutional Neural Network

Description:

This work presents a wheat variety classification system based on a Spatial Deep Convolutional Neural Network (SD-CNN). The proposed model is designed to automatically learn discriminative spatial patterns from wheat kernel characteristics and accurately classify different wheat varieties.

Seven representative features—area, perimeter, compactness, kernel length, kernel width, asymmetry coefficient, and kernel groove length—are used as inputs to the network. These features capture the spatial and morphological properties of wheat kernels that are crucial for distinguishing between varieties.

The Spatial Deep CNN consists of multiple convolutional layers that extract high-level spatial representations from the input feature space, followed by fully connected layers for classification. During training, the network learns complex non-linear relationships between the input features and the corresponding wheat variety labels. The learned model is then employed to predict the variety of unseen wheat samples.

Model performance is evaluated using standard classification metrics such as accuracy, precision, recall, and F1-score, demonstrating the effectiveness of the proposed approach in wheat variety classification. The results confirm that the Spatial Deep CNN provides a robust and efficient solution for automated agricultural decision-support systems.

Input  : wheat kernel features

Output  : wheat variety

Tags :

#WheatVarietyClassification, #SpatialDeepCNN, #DeepLearning, #ConvolutionalNeuralNetwork, #SeedMorphology, #GrainQualityAnalysis, #AgriculturalAI, #AgriTech, #AIInAgriculture, #CropScience, #ComputerVision, #PatternRecognition, #FeatureBasedClassification, #SupervisedLearning, #AgroInformatics, #SmartFarming, #FoodSecurity, #IntelligentAgriculture

References:

1, Singh, R., Patel, N., and Verma, S. (2020). Wheat variety classification using convolutional neural networks based on grain morphological features. International Journal of Advanced Computer Science and Applications (IJACSA), IEEE.

2, Sharma, P., Mehta, K., and Joshi, R. (2021). Deep learning–based classification of wheat seed varieties using spatial feature extraction. International Journal of Intelligent Systems and Applications in Agriculture.