Intelligent processing method, device and equipment for video stream and natural language in navigation

Abstract

The invention discloses an intelligent processing method, device and equipment for a video stream and a natural language in navigation. The method comprises the following steps: forming a natural language image knowledge base; extracting decision languages and semantics in natural languages; segmenting a decision language part and a non-decision language part; recognizing and translating the natural language of the non-decision language part into an image frame sequence; the image frame sequence forms a video stream or the image frame sequence and the video stream are matched and fused; establishing a space-time polar coordinate system in the video stream frame image; analyzing a dynamic image and a non-dynamic image under the space-time polar coordinate system; establishing a decision tree structured model mechanism, and traversing the decision tree to mark multi-dimensional parameters; an image content language matrix and a video stream content language matrix are formed, and intelligent processing from a video stream to a natural language is completed; performing navigation semantic understanding according to a natural language; the method has the advantages that the mutual mapping, matching and fusion efficiency is high, the accuracy, real-time performance and safety of intelligent navigation are improved, and the experience degree is improved.

Description

technical field [0001] The invention relates to the technical field of intelligent processing of video streams and natural languages, in particular to a method, device and equipment for intelligent processing of video streams and natural languages ​​in navigation. Background technique [0002] At present, with the vigorous development of the air, sky, earth and sea information network, various intelligent mobile terminals are rapidly changing people's lives, such as planets, satellites, aircraft, ships, automobiles, mobile phones, hot air balloons, and various types of robots for military and civilian use. in full swing. "Internet + robot" is gradually being upgraded, and the demand is very strong. [0003] At the same time, people's requirements for intelligent navigation of various intelligent mobile terminals are constantly improving, and automatic driving and obstacle avoidance technologies have received more and more attention. Intelligent processing of video streams a...

AI Technical Summary

Problems solved by technology

[0004] In the existing technology, more literature researches focus on the detection, segmentation and recognition of video images and more literature researches focus on natural language processing. The relative research on intelligent processing between video streams and natural language is still very immature. , the research on intelligent processing of video stream and natural language in navigation is forward-looking, basic and challenging, and it is the frontier of international research, and related research is still lacking.

Images