Person Sensor-Aided Scene Recognition and Understanding Based on CG Technology

Abstract

Person sensor-aided scene recognition and detailed understanding based on CG Technology is studied in this paper. Researchers have been working on enabling computers to quickly locate and recognize textual information in images of natural scenes. To achieve this target, we analyze the CG tool, computer graphics is used as the science that uses mathematical algorithms to transform two-dimensional or three-dimensional graphics into various forms of the grids in a computer display. Applications in this field mainly refer to the effective control, generation, storage, and communication of multimedia information by means of the computer control or then grassroots information system. The optimization of the core network structure not only allows the network to fully extract low-level features, but also reduces the interference of complex scenes on object recognition after two capsule layers have undergone two-layer dynamic routing and screening decisions. This paper proposes the novel framework for the optimal model. The experiment proves the performance.