LSCC-Net: Convolutional Neural Network for Coated Corn Seed Variety Recognition

To achieve the recognition of coated corn seed varieties and address the issue of high complexity in deep learning models, which makes them difficult to adapt to hardware devices with limited computing power, this paper proposes a lightweight convolutional neural network method for recognizing coated corn seed varieties, named LSCC-Net. A dual-path structure DFE Block is proposed as the feature extraction module of LSCC-Net, enabling efficient extraction of feature information with a small number of parameters. Experimental results show that the LSCC-Net model achieves an accuracy of 97.9<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> on the coated corn seed dataset with only 0.16M parameters. The LSCC-Net algorithm is characterized by its lightweight nature and high accuracy, making it suitable for deployment on hardware devices with limited computing power. It provides a theoretical foundation for the automated detection of corn seeds.