Maize Haploid Identification via LSTM-CNN and Hyperspectral Imaging Technology

Abstract

Accurate and fast identification of seed cultivars is crucial to plant breeding, with accelerating breeding of new products and increasing its quality. In our study, the first attempt to design a high-accurate identification model of maize haploid seeds from diploid ones based on optimum waveband selection of the LSTM-CNN algorithm is realized via deep learning and hyperspectral imaging technology, with accuracy reaching 97% in the determining optimum waveband of 1367.6-1526.4nm. The verification of testing another cultivar achieved an accuracy of 93% in the same waveband. The model collected images of 256 wavebands of seeds in the spectral region of 862.9-1704.2nm. The high-noise waveband intervals were found and deleted by the LSTM. The optimum-data waveband intervals were determined by CNN's waveband-based detection. The optimum sample set for network training only accounted for 1/5 of total sample data. The accuracy was significantly higher than the full-waveband modeling or modeling of any other wavebands. Our study demonstrates that the proposed model has outstanding effect on maize haploid identification and it could be generalized to some extent.

Accelerated Generation of Elite Inbreds in Maize Using Doubled Haploid Technology

“…The effectiveness of the oilbased identification technique, on the other hand, is dependent on a large difference in oil content between source germplasm and inducer, since a little difference would result in a higher number of false positives and false negatives [21,22]. Automating the process of haploid identification would be a cost-effective and practical solution since it would considerably cut the cost of wages for those participating in the haploid identification process [23]. Several mechanical approaches have been altered based on R1-nj marker expression on embryo and endosperm employing multispectral, hyperspectral, and fluorescence imaging techniques (Figure 3).…”

“…Several mechanical approaches have been altered based on R1-nj marker expression on embryo and endosperm employing multispectral, hyperspectral, and fluorescence imaging techniques (Figure 3). In this case, an imagining-based automated approach powered by machine learning and deep learning understanding might Accelerated generation of elite inbreds in maize using doubled haploid technology DOI: http://dx.doi.org/10.5772/intechopen.105824 be a feasible option since it decreases the time and effort required to identify haploids [23,24]. As a result, numerous approaches for haploid identification employing in-vivo HI are available.…”