System and method for early warning of water pollution based on fish multi-sample statistics

Abstract

The invention discloses a system for early warning of water pollution based on fish multi-sample statistics. The system comprises a multi-part fish tank, an image acquisition device and a processor, wherein the multi-part fish tank is divided into at least three parts and used for accommodating circulating raw water to be monitored, and the parts of the fish tank can be used for accommodating small fishes with the same quantity and type; the image acquisition device is used for acquiring continuous image sequence frames of the small fishes in the multi-part fish tank in real time; and the processor is used for dividing the sequence frames according to a set fish tanks alternating strategy to obtain monitored and referenced image sequence frames corresponding to monitored and referenced fish tanks, respectively processing the monitored and referenced image frames to obtain motion parameters of monitored and referenced small fishes, respectively judging whether the monitored fish tank is in an alarm range or not and the small fishes in the referenced fish tank are in an abnormal state or not, and confirming that the raw water is polluted when the monitored fish tank is in the alarm range and the small fishes in the referenced fish tank are in the abnormal state. The invention further discloses a corresponding method for early warning. Due to the adoption of the system and the method for early warning of water pollution, the accuracy for judging whether the raw water is polluted or not is greatly improved.

Description

[0001] technical field [0002] The invention relates to a water quality pollution early warning system and method, in particular to a water quality pollution early warning system and early warning method based on fish multi-sample statistics. [0003] Background technique [0004] The invention patent with the patent number of 200910038318.8 discloses a water pollution early warning method. The method collects image sequence frames of small fish in a fish tank equipped with raw water to be monitored, and performs smoothing, threshold segmentation and removal of the collected image sequence frames. Noise processing, calculating motion parameters of the small fish according to the image sequence frames after denoising processing, and determining that the detected raw water quality is polluted when the calculated motion parameters exceed a predetermined range. [0005] However, this early warning method only judges the water quality by detecting the movement behavior of a gro...

AI Technical Summary

Problems solved by technology

[0005] However, this early warning method only judges the water quality by detecting the movement behavior of a group of fish, the judgment factor is single, and the accuracy is low

For example, behavioral deviations such as chasing between fish schools or factors such as fish fatigue caused by continuous long-term work of fish schools will affect the movement parameters of fish schools, which will lead to large errors in calculated movement parameters, thus affecting whether the raw water quality is affected. Judgment of pollution

Images