Maize haploid seed selection method based on CNN-SVM

Yifeng Zhu; Xingyu Zhou; Lanzhen Yao; Xiaofeng An

doi:10.1117/12.2685536

28 July 2023 Maize haploid seed selection method based on CNN-SVM

Yifeng Zhu, Xingyu Zhou, Lanzhen Yao, Xiaofeng An

Proceedings Volume 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023); 127160U (2023) https://doi.org/10.1117/12.2685536
Event: Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 2023, Xi'an, China

Abstract

The haploid breeding technology of maize can shorten the breeding cycle and is an important technology for modern crop improvement. However, selection of maize haploid seeds is often done manually, resulting in loss of time and labor. It is of scientific value to study the selection algorithm of maize haploid seeds with high accuracy and strong generalization ability. In this paper, CNN and SVM were combined, and CNN-SVM model was used to classify maize seeds. The optimal CNN model is obtained by adjusting the number of convolution layers through experiments. In the training process of SVM, the cuckoo search algorithm is used to optimize the value of hyperparameter C and hyperparameter gamma, so as to improve the training efficiency and classification performance of SVM. The performance of CNN-SVM model was compared with CNN-KNN, CNN, HOG-KNN, HOG-SVM, SURF-KNN, SURF-SVM, Pixel histogram-KNN, Pixel histogram-SVM. Experimental results show that the CNN-SVM model is superior to other models, and the accuracy of maize seeds classification is 98.5%.

Citation Download Citation

Yifeng Zhu, Xingyu Zhou, Lanzhen Yao, and Xiaofeng An "Maize haploid seed selection method based on CNN-SVM", Proc. SPIE 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 127160U (28 July 2023); https://doi.org/10.1117/12.2685536

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available