A novel pest classification method based on the compressed sensing

The classification of stored grain pests based on the computer vision technology is studied in this paper. Combining with compressed sensing sparse representation theory, a novel stored grain pests classification model which meets the RIP condition is proposed. First, the grain pests based on sparse representation classification model is built, and then a condition which satisfied the RIP pest classification model is derived, and the RIP of the model as well as compressed sensing reconstruction model of equivalence is proved. Simulation results show that: the proposed model is superior to the existing pest classification model, in comparison with other classification algorithm, the proposed classification method still achieved good classification results.