State judgment system and judgment method for AI recognition based on grouping patterns

Abstract

The invention discloses a state judgment system and judgment method for AI recognition based on grouping patterns. The state judgment system comprises standard grouping patterns, a grouping learning set, an AI recognition device, a state verification platform and a logic management platform. The standard grouping pattern is formed by grouping a plurality of standard patterns, and each group comprises a plurality of patterns; wherein the grouping learning set is used for carrying out AI recognition learning according to each standard grouping pattern; the AI recognition device is used for recognizing a plurality of specific grouping patterns; and the state verification platform is used for receiving the state report of the AI recognition device with the specified number, obtaining the logicstate of the equipment with the corresponding number from the logic management platform and carrying out verification. The judgment method of the system is simple, the recognition speed is high, thehardware facility cost is low, installation, operation and maintenance of the AI recognition equipment are simplified to a great extent, and therefore it needs to be ensured that the correct key application operation environment is safer and more reliable.

Description

technical field [0001] The invention relates to the field of recognition technology, in particular to a state judgment system and a judgment method for AI recognition based on grouping patterns. Background technique [0002] With the advancement of technology, a large number of devices and systems with AI recognition functions have appeared. However, these devices generally need to process a large amount of video data, adopt general-purpose recognition algorithms, and require high computing power, resulting in high requirements for hardware. Moreover, due to the uncertainty of the object to be recognized, such as a face or an object, the processing logic is complex, and it is difficult to perform accurate recognition. It can only be similarity recognition, that is, a high misrecognition rate must be tolerated. [0003] There is a class of occasions that must be accurately identified, such as the selection of railway switches, the selection of buttons / actions in the main con...

AI Technical Summary

Problems solved by technology

[0005] (1) Fatigue, temper tantrum, mental illness, etc. of the staff will lead to the occurrence of major liability accidents;

[0006] (2) In the event of an emergency, since the staff has never encountered the incident, they may make mistakes in handling actions due to nervousness, resulting in accidents;

[0007] (3) In the case of harsh environment, for example, when receiving interference from light irradiation and environmental shaking, the staff is prone to misoperation;

[0008] (4) In many occasions, the handling of events often needs to be decomposed into a combination of several actions. Even well-trained staff may have problems due to wrong judgments when performing a certain step

Images