Intelligent identification system and method for worker edge falling facing dangerous omen reasoning

Abstract

The invention discloses a worker limb falling intelligent identification system and method oriented to risk omen reasoning, and the method comprises the steps: extracting features of key objects of limb falling scenes, and determining the spatial range of each limb dangerous scene; knowledge structured combination is carried out according to the subject class, the attribute class, the object class, the position class and the behavior state class, and a limb area insurance mega knowledge graph is constructed through a graph database; performing visual identification on the edge falling dangerous area and the characteristics of the behavior state of the worker; and evaluating the safety state of the worker limb operation by utilizing graph database reasoning according to a visual identification result. According to the method, in combination with the behavior state of a worker, the constructed edge region risk omen knowledge rule is utilized, and the edge risk omen event is identified on the basis of intelligent visual detection. The system is high in intelligent level, and has high expansibility and practical value.

Description

technical field [0001] The invention relates to the technical field of semantic understanding of computer vision, including deep learning technology and semantic reasoning technology, and in particular to an intelligent recognition system and method for a worker's edge fall oriented to dangerous reasoning. Background technique [0002] The construction industry is an important industry supporting social and economic development, and it is also a typical high-risk industry. Among the many types of safety accidents, falling from heights is the most important type of accidents, accounting for more than half of the accidents. In various high-altitude operation scenarios, climbing, suspension, platform, and cross operations are difficult, and experienced workers are usually selected and equipped with reliable safety equipment, and the probability of falling accidents from heights is very small. However, falling from the edge operation, which is not very difficult, has become the...

AI Technical Summary

Problems solved by technology

However, if we only rely on the superpower of computer vision in the field of image perception and ignore the semantic relationship between perceived objects, there will be a "semantic gap" between the underlying features and the high-level semantics, that is, the perception results will have a high false positive rate.

Images