Graph division method adaptive to visual law based on neural network

Abstract

The invention relates to the technical field of neural networks, and discloses a neural network-based visual law-adaptive graph division method, which comprises the following specific steps of S1, building a neural network model of a divided graph; s2, preparing a material for training the sub-graph neural network model; s3, training a neural network model; and S4, compressing the neural network model of the sub-graph and testing. According to the neural network-based graph division method adapting to the visual law, the visual effect is judged through the appropriate crowd and the original graph data, the feedback result is transmitted to the deep neural network graph division model, the deep neural network model is obtained through large-scale data training, the visual induction of the crowd can be effectively improved, and the user experience is improved. A common image is split into a plurality of images through the model, the plurality of images do not have significance but can show the visual effect of an original image in a mode conforming to a visual rule, the original image is prevented from being copied through a simple means, and the visual effect of the image can be transmitted out under the condition that the original image is not transmitted.

Description

technical field [0001] The invention relates to the technical field of neural networks, in particular to a method for dividing graphs based on neural networks to adapt to visual laws. Background technique [0002] Neural network is an algorithmic mathematical model that imitates the behavioral characteristics of animal neural network and performs distributed parallel information processing. This kind of network relies on the complexity of the system and adjusts the interconnected relationship between a large number of internal nodes to achieve the purpose of processing information. , its main task is to build a practical artificial neural network model according to the principle of biological neural network and the needs of practical applications, design corresponding learning algorithms, simulate certain intelligent activities of the human brain, and then implement it technically to solve practical problems. question. [0003] Most of the existing computer image segmentati...

AI Technical Summary

Problems solved by technology

[0003] Most of the existing computer image segmentation methods are based on color values ​​and image regions. There are problems such as the stretching of the original image in time and poor visual sensing, and the original image is easily copied during use. To solve the above problems, we propose a neural network-based segmentation method that adapts to visual laws

Images