NEURAL NETWORK-BASED TARGET BACKGROUND NOISE SUPPRESSION METHOD AND DEVICE

摘要

A neural network-based dense target background noise suppression method and a device, the method comprising: acquiring an image and determining multiple densely distributed targets to be segmented; individually performing external frame segmentation of said targets to be segmented, obtaining external frames corresponding to the targets to be segmented; performing hierarchical sorting processing of the targets to be segmented; detecting angles of rotation of the targets to be segmented; performing external frame segmentation of the targets to be segmented; outputting the targets to be segmented. A computer-readable medium having stored thereon a computer-readable instruction, the computer-readable instruction being capable of causing a processor to implement the described method when executed by the processor. The device comprises: a processor; a computer-readable medium used to store a computer-readable instruction, such that when the computer-readable instruction is executed by the processor, this causes the processor to implement the described method.