Water environment AI target object identification method and system thereof

Abstract

The invention discloses a water environment AI target identification method and system. The method comprises the following steps: 1, preprocessing each frame of image of a monitored riverway surface to generate preprocessed image data; 2, adopting a k-means clustering algorithm to perform clustering processing on image data to generate a clustering result, and dividing class value into three classes which represent people, vehicles, ships and floaters in the river surface respectively; 3, performing image recognition on preprocessed image data, determining all recognizable lines in an image, and generating sample lines; and 4, performing fitting processing on the sample line to determine the boundary position of the river channel and the shore. The invention has the beneficial effects thatthe problem that a large amount of manpower needs to be input for current river long system river channel patrol is solved. The functions of real-time monitoring, automatic alarming and automatic classified pushing are achieved, and the problems of missing alarming, false alarming and untimely problem reporting are solved; the problem that traditional border-crossing alarms cannot be pushed in aclassified mode is solved.

Description

technical field [0001] The present invention relates to the field of AI image pattern recognition, mainly relates to the field of specific target recognition on rivers, lakes or banks, specifically a water environment AI target recognition method and system thereof. Background technique [0002] my country has a vast territory, rich in rivers, reservoirs, lakes, and sea areas. Among them, water conservancy informatization, safe and smooth waterway traffic, fishery management, and safe operation of ports all require modern monitoring facilities and means. Long-distance wireless video monitoring systems are installed in every important river basin in my country. During the rainy season and flood season, the surface of rivers and lakes or banks often changes, so it is necessary to monitor the surface of rivers, lakes or banks to keep abreast of changes in water surface, detect dangerous situations in time, and To deal with this in a timely manner, to achieve manual management of...

AI Technical Summary

Problems solved by technology

[0003] The existing river channel monitoring method, one is to monitor by manpower, but it needs to invest a lot of manpower, and in the rainy season or flood season, there are potential safety hazards in manual inspection; the other is to set up monitoring equipment separately at each monitoring point, and monitor for a period of time , through the data generated by video recognition products to analyze the information on the surface of rivers and lakes or banks, but the existing video recognition products cannot intelligently identify floating objects on the surface of rivers and lakes or specific targets on the river bank, which is prone to false positives, Missing or untimely reporting, it is impossible to comprehensively monitor the real-time status of rivers, lakes or banks

Images