Hyper spectral imaging technology for seed quality identification

Agriculture is the backbone for any Nation. The quality of agriculture is depends on the amount of production. Seed is an elementary contribution in agriculture which is a supreme and crucial input for best crop production, one of the methods to surge the yield through the plantation of quality seed. It is the high time for reconsideration of the farming processes in the agriculture specifically the development of the seed in a low cost rate and the usage of a reduced amount of fertilizers. Hyperspectral imaging (HSI) technology is a procedure which makes use of different classification models such as Deep convolution Neural network (DCNN), Partial least squares regression (PLSR) and Deep Forest (DF) for the prediction of growth rate of the seed. The HSI system covering the spectral range of 948-2494 nm. The Deep Forest (DF) model to elevate spectral features in the seeds. In this study the hyperspectral images of seeds were used to get the objects. This paper represents a topo spectral view on using the various machine learning algorithms that evades the development of the seed in comparison with various machine learning algorithms.