Study on Multiple Varieties of Maize Haploid Qualitative Identification Based on Deep Belief Network
YU Yun-hua1,2, LI Hao-guang1,2, SHEN Xue-feng1,2, PANG Yan1
1. Shengli College,China University of Petroleum,Dongying 257061,China
2. College of Information and Control Engineering,China University of Petroleum,Dongying 257061,China
Abstract:Haploid breeding technology is a new method for maize breeding, which can effectively shorten the cycle of homozygous lines and improve the breeding efficiency. The technology needs to select enough haploid grains first, and the haploid grains only account for 0.05%~0.1% of the mixed grains without artificial intervention. Even with the biological induction technology, the number of haploid grains is generally less than 10%. High-speed and accurate identification of haploid grains can meet the!needs of engineering breeding. However, molecular biology and morphological identification methods commonly used in practical work are time-consuming, costly and destroying samples. It is difficult to select Maize Haploid grains efficiently and accurately. Relevant studies have proved that there are obviousoil content differences between haploid and diploid of high-oil maize. At present, low-field nuclear magnetic resonance technology can be used to detect oil content of maize and identify haploid according to its oil content. However, nuclear magnetic resonance (NMR) instrument has some weaknesses, such as high price, difficult maintenance, slow speed and low efficiency. It takes 4 seconds for each single-grain sorting. It cannot meet the needs of large number identification for engineering breeding. Using VIAVI near infrared spectrometer (NIRS) can achieve the detection speed of 0.25 seconds for each maize. The NIR technology is faster, cheaper and easier to maintain. The NIR identification method can replace the method of NMR. Qualitative identification of haploid by NIRS has achieved some results, but currently there are relatively few maize varieties collected in the study. The study only establishes models for haploid of one variety, and classifies haploid of that variety. There are no studies on identification of multiple hybrid haploids at home and abroad, but engineering breeding urgently needs a method to identify multiple varieties of maize haploids. In this paper, a method for identifying haploids based on deep belief network is proposed. DBN is a multi-layer deep neural network. Each layer is composed of a restricted Boltzmann mechanism. By using layer-by-layer training strategy, the problem that traditional neural network training methods are not suitable for multi-layer training can be solved. The comparative experimental results show that the identification model of multiple varietieshaploid established by DBN method has high classification performance and can meet the requirements of maize engineering breeding accuracy.
Key words:Deep belief net; Haploid; Near infrared spectroscopy; Qualitative identification
于云华,李浩光,沈学锋,逄 燕. 基于深度信念网络的多品种玉米单倍体定性鉴别方法研究[J]. 光谱学与光谱分析, 2019, 39(03): 905-909.
YU Yun-hua, LI Hao-guang, SHEN Xue-feng, PANG Yan. Study on Multiple Varieties of Maize Haploid Qualitative Identification Based on Deep Belief Network. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(03): 905-909.
[1] CHEN Shao-jiang,LI Liang, LI Hao-chuan(陈绍江, 黎 亮, 李浩川). Maize Haploid Breeding Technology(玉米单倍体育种技术). Beijing: China Agricultural University Press(北京:中国农业大学出版社), 2012.
[2] Li Haochuan, Yang Jiwei, Qu Yanzhi, et al. Journal of Maize Sciences,2015(1):73.
[3] Melchinger A E, Schipprack W, Friedrich Utz H, et al. Crop Sci., 2014, 54(4): 1497.
[4] Albrecht E Melchinger, Markus Winter, Mi Xuefei, et al. Crop Science, 2015, 55(5~6): 1.
[5] Albrecht E Melchinger, Wolfgang Schipprack, Mi Xuefei, et al. Crop Sci.,2015, 55(1~2): 188.
[6] Dong X, Xu X, Li L, et al. Molecular Breeding, 2014, 34(3): 1147.
[7] Dong X, Xu X, Miao J, et al. Theoretical and Applied Genetics, 2013, 126(7): 1713.
[8] Liu Jin, Guo Tingting, Yang Peiqiang, et al. Transactions of the Chinese Society of Agricultural Engineering,2012, 28 (Suppl. 2): 233.
[9] QIN Hong, MA Jing-yi, CHEN Shao-jiang, et al (覃 鸿,马竞一,陈绍江,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2015, 25(11):1807.
[10] QIN Hong, MA Jing-yi, CHEN Shao-jiang, et al(覃 鸿,马竞一,陈绍江,等). Infrared Technology(红外技术),2015, 1(37): 78.
[11] LI Hao-guang, LI Wei-jun, QIN Hong, et al(李浩光,李卫军,覃 鸿,等). Transactions of The Chinese Society of Agricultural Machinery(农业机械学报),2016, 47(6): 259.
[12] Hinton G, Deng L, Yu D, et al. IEEE Signal Processing Magazine, 2012, 29(6): 82.
[13] Stellan Ohlsson. Deep Learning. Cambridge University Press, 2011.
[14] Shoujue W, Jiangliang L. Neurocomputing, 2005, 67: 9.