Printing typesetting system and method based on newspaper layout AI automatic typesetting design

Abstract

The invention relates to a printing typesetting system and method based on newspaper AI automatic typesetting design, particularly relates to an intelligent typesetting design method for publishing newspapers and fusing layout design principles, in particular to the field of artificial intelligence and big data, which comprises the following steps of: providing a printing newspaper typesetting design software system, a layout data center, a graphic design rule and enabling a convolution kernel of a convolution layer to convolve initialized images and texts; and extracting most representative feature points (layout titles, fonts, word sizes and pictures) in the images and texts; correcting automatic layout design and survival training sets by using newspaper layout design rules; performinglayout structure and color classification and structural layout and color distribution AI automatic typesetting design on layout titles, fonts, word sizes and pictures to be printed through the trained layout model in combination with the printed newspaper layout to generate a printed output newspaper design layout; entering a printing link and performing batch production by using a printer.

Description

technical field [0001] The invention belongs to a printing typesetting system and method based on AI automatic typesetting design of newspaper layout, especially about newspaper publishing, an intelligent typesetting design method integrating layout design principles, specifically related to artificial intelligence, big data, and its rule intelligent typesetting design and Artificial intelligence algorithm technology field. For newspaper publishing, AI automatic typesetting design must not only solve the problem of conveying visual art, but also consider how information can be accurately conveyed to readers and achieve the highest utilization rate. In particular, it involves calculating the word space difference for each of the candidate templates through AI deep reinforcement learning, comparing the difference between different candidate templates, and finally selecting the template with the smallest difference as the best template. Background technique [0002] Typography...

AI Technical Summary

Problems solved by technology

However, the scheduling strategy of the list heuristic scheduling algorithm uses the earliest completion time of a single task as the basis, making each scheduling a greedy strategy, and it is difficult to obtain globally optimized scheduling

The scheduling strategy of the scheduling algorithm based on task replication uses multiple identical tasks to be executed on different processors. Although the communication overhead is reduced, the number of task executions is increased, which greatly increases the consumption of computing resources.

The scheduling strategy of the random search algorithm based on genetics and evolution requires a large number of iterative calculations. The iterative process of random search takes too long to execute and cannot meet the task scheduling scenarios with high real-time requirements.

Images