期刊论文

中文

中国粮油学报

1003-0174

2024

39

7

181-188

10.20048/j.cnki.issn.1003-0174.000691

61272278:国家自然科学基金 2018ABA099:湖北省重大科技专项基金项目 2018A01038:天诚汇智创新促教基金项目

本校为第一机构

针对现有的大米外观品质检测方法存在识别准确率较低的问题,本研究提出一种结合高效通道注意力机制(Efficient Channel Attention,ECA)的改进CenterNet模型(SEP-CenterNet)用于...展开更多 针对现有的大米外观品质检测方法存在识别准确率较低的问题,本研究提出一种结合高效通道注意力机制(Efficient Channel Attention,ECA)的改进CenterNet模型(SEP-CenterNet)用于大米外观品质检测。首先,使用图像采集设备获取碎米、整米和黄米图像,为防止由于数据集较小出现过拟合现状,对图像进行旋转、翻转等方式扩充数据集;然后将引入空间金字塔池化(Spatial Pyramid Pooling,SPP)的GhostNet作为CenterNet的...

In this research,an improved CenterNet model(SEP-CenterNet)combining Efficient Channel Attention(ECA)mechanism for rice appearance quality detection was proposed,aiming to solve the problem of low recognition accuracy in existing detection methods.Firstly,images of brok...MORE In this research,an improved CenterNet model(SEP-CenterNet)combining Efficient Channel Attention(ECA)mechanism for rice appearance quality detection was proposed,aiming to solve the problem of low recognition accuracy in existing detection methods.Firstly,images of broken rice,whole rice,and yellow rice were obtained usi...