Animal long-time posture recognition system based on infrared image and wearable optical fiber

Abstract

The invention discloses an animal long-time posture recognition system based on an infrared image and a wearable optical fiber. The system comprises an attention mechanism module M1, a local posture recognition model training module M2, a local posture recognition prediction module M3, and a global posture restoration module M4. According to the invention, high-precision identification of the keynode of the animal behavioral infrared low-resolution video wearing the optical fiber equipment is realized, the time length of manual verification is greatly reduced, the influence of animal curling,rotation, shooting illumination fluctuation and the like is solved, and the system can be widely applied to the field of brain imaging.

Description

[0001] The invention relates to the field of image processing, in particular to an animal long-term posture recognition system based on infrared images and wearable optical fibers. Background technique [0002] The behavior of neurons is closely related to animal activity. Quantitative analysis of animal behavior is often required to explore how complex networks of neurons connect and interact for specific behaviors. [0003] Under the current high-frequency acquisition conditions of big data, the efficiency of manual processing is time-consuming and error-prone, but due to the dynamics and complexity of natural animal behavior and the interference of surrounding factors, automatic identification is a great challenge. Animals wearing miniature microscopes have optical fibers on their heads during the laser signal transmission process. Due to the occlusion and shaking of the optical fibers, behavioral analysis will pose a greater challenge. At the same time, during the long-te...

AI Technical Summary

Problems solved by technology

[0003] Under the current high-frequency acquisition conditions of big data, the efficiency of manual processing is time-consuming and error-prone, but due to the dynamics and complexity of natural animal behavior and the interference of surrounding factors, automatic identification is a great challenge

Animals wearing miniature microscopes have optical fibers on their heads during the laser signal transmission process. Due to the occlusion and shaking of optical fibers, behavioral analysis will pose a greater challenge

At the same time, during the long-term experimental recording process, environmental conditions such as image lighting often fluctuate non-negligibly, which interferes with the stability of features.

In addition, in order not to interfere with the excitation imaging of wearable fluorescence microscopes, infrared cameras are often used to record the behaviors recorded, while infrared images often have low contrast, poor resolution, and blurred texture information, which significantly increases the automatic recognition of experimental animal behavior difficulty

[0004] Previous attempts to use physical markers will interfere with the natural behavior of animals; some have also tried to use depth cameras to locate animal behavior in natural space, but these technologies are sensitive to imaging conditions and scene changes; traditional gesture recognition methods rely on The extraction and matching of features requires high feature stability, so it is not suitable for complex moving animal behavior recognition

Images