基于深度学习的高光谱特征玉米品种纯度识别方法  

A Deep Learning Based Method for Identifying the Purity of Corn Seed Varieties with Hyperspectral Features

作　　者：贺文文 田承华 HE Wenwen;TIAN Chenghua(Sorghum Institute,Shanxi Agricultural University,Jinzhong,Shanxi 030600,China)

机构地区：[1]山西农业大学高粱研究所,山西晋中030600

出　　处：《作物研究》2024年第4期272-277,313,共7页Crop Research

基　　金：吕梁市校地合作科技成果转化推广专项计划(2023XDHZ14)。

摘　　要：为提升玉米种子的识别精度,提高玉米的产量和质量,基于高光谱成像和深度学习技术构建了一种包含数据预处理、图像分割、基于改进卷积神经网络(convolutional neural network,CNN)的高光谱玉米种子纯度识别方法。研究结果表明,改进的CNN在提高训练性能方面效果显著,其准确度、精确度、召回率和F1分数等指标均优于传统机器学习和其他深度学习方法。In order to improve the recognition accuracy of corn seeds and improve the yield and quality of corn,a method based on hyperspectral imaging and deep learning technology including data preprocessing,image segmentation,and improved convolutional neural network(CNN)based on hyperspectral corn seed purity identification method.The research results show that the improved CNN has significant effects in improving training performance,and its accuracy,precision,recall rate and F1 score are superior to traditional machine learning and other deep learning methods.

关 键 词：玉米 种子 纯度识别 高光谱 数据预处理 深度学习 

分 类 号：S123[农业科学—农业基础科学] S513.024