An intelligent monitoring method for wild animals based on multi-dimensional information fusion

Abstract

The invention discloses an intelligent monitoring method for wild animals based on multi-dimensional information fusion, which includes: using a drone equipped with a dual-light coaxial camera to obtain a fusion image of a visible light image and an infrared thermal image of a wild animal according to a preset route; The human-machine automatic navigation algorithm tracks the direction of the wild animal footprints to obtain the footprint chain image; the wild animal fusion image and the footprint chain image are spliced ​​separately to obtain the spliced ​​image; the vibration sensor network is pre-embedded in the monitoring area, and the vibration signal collected by the vibration sensor network Perform real-time online identification and classification, and upload the classification results to the monitoring terminal; use the pre-trained dual-channel network recognition model to analyze the spliced ​​images, obtain the classification results of wild animal species in the monitoring area, and upload them to the monitoring terminal. The invention can carry out multi-dimensional and all-weather monitoring and identification of wild animals in wild animal habitats, greatly improving identification accuracy and work efficiency.

Description

technical field [0001] The invention relates to the technical field of wild animal monitoring, and more specifically relates to an intelligent monitoring method for wild animals based on multi-dimensional information fusion. Background technique [0002] Wild animals are one of the important members of the ecosystem. They are not only closely related to the maintenance of ecological balance, but also inseparable from human life and social development. With the improvement of productivity level, it has brought increasingly serious ecological and environmental problems, causing many wild animals to be on the verge of extinction. In order to protect endangered wild animals, the monitoring work should not be underestimated, and the quality of the monitoring work is directly related to whether the real-time protection of wild animals can be implemented. [0003] At present, infrared cameras or drones equipped with visible light cameras are mainly used to monitor wild animals. I...

AI Technical Summary

Problems solved by technology

However, visible light sensors are sensitive to changes in light, visibility at night, and other environmental impacts, making it impossible to monitor wild animals around the clock

[0004] At the same time, infrared cameras or UAVs equipped with visible light cameras cannot conduct comprehensive records and intelligent analysis of wild animals' weight, approximate location and footprint chain.

[0005] In recent years, deep learning has made major breakthroughs in the field of artificial intelligence, and has achieved great success in the fields of speech recognition, image recognition, and video analysis. When monitoring wild animals, the image data is vulnerable to the behavior of the animals themselves when shooting in the wild. , the influence of the surrounding environment and climate, poor imaging conditions, resulting in low recognition accuracy, especially in natural scenes, the intervention of nonlinear factors such as lighting, the occlusion and angle problems caused by the non-cooperative research targets of wild animals, making Some traditional algorithms are difficult to apply directly to this problem, and the recognition accuracy is low

Images