Visual perception type water quality early warning system and method using olfaction signal enhancement

Abstract

The invention provides a visual perception type water quality early warning system and method using olfaction signal enhancement. The method comprises the steps of: carrying out continuous monitoring on the characteristics of a target fish behavior through a visual model, and judging whether the target fish behavior is abnormal or not in real time; activating further detection when a detection result contains an abnormal stress behavior, and performing olfaction detection on the fish water environment with the stress behavior; and correcting a data set by using an abnormal water quality component result of the olfaction detection and the calculated characteristic value of the target fish behavior, and storing the data set in an experience pool, so as to continuously correct the behavior abnormality problem of the target fish caused by harmful compounds contained in the water. The characteristics of the target fish behavior monitored by the visual model continuously fit the behavior characteristic curve of fishes in a real water environment. According to the method, the perception ability of the visual model in vision is continuously enhanced by using the olfaction signal, the complementarity of heterogeneous data signals is fully played, the accuracy, the real-time performance and the high efficiency of the water quality early warning model are improved, and the consumption of time, space and human resources is reduced.

Description

technical field [0001] The invention relates to the technical field of signal processing, in particular to a visual perception type water quality early warning system and method utilizing olfactory signal enhancement. Background technique [0002] Biologically, visual and olfactory signals often have a synergistic effect on the perception capabilities of living things. For example, biologists found that fruit flies not only integrated the functions of visual and olfactory signals in the process of food searching, but also concluded that fruit flies that had been preprocessed with olfactory signals also enhanced their ability to capture visual signals. Therefore, inspired by the fact that olfactory signals can enhance visual perception in biology, the present invention proposes a visual perception water quality early warning method applied in the field of water quality detection by utilizing olfactory signals. [0003] In the field of water quality detection, although the tr...

AI Technical Summary

Problems solved by technology

[0003] In the field of water quality detection, although the traditional method of using olfactory sensors to detect water quality-related physical and chemical parameters has high accuracy, it also has the disadvantages of time-consuming, human resources, and poor detection timeliness; while using computer vision Although the method of biological water quality detection in related technologies in the field has the characteristics of high detection timeliness, it is not accurate to judge the water quality only based on biological behavior characteristics

Images