Research on motion target detection based on infrared biomimetic compound eye camera

Area of Science:

• Optics and Vision Science
• Biomimetics
• Computer Vision

Background:

• Cooled mid-wave infrared biomimetic compound eye cameras offer broad applications in industrial inspection, military, and security.
• Existing motion target detection algorithms struggle with low-resolution individual eyes and wide fields of view, leading to missed targets.
• The limitations necessitate advanced algorithms for effective motion detection in these systems.

Purpose of the Study:

• To develop an improved motion detection algorithm for biomimetic compound eye cameras.
• To enhance the detection of moving targets while suppressing stationary backgrounds.
• To reduce missed detection rates and improve the accuracy and stability of results.

Main Methods:

• Proposed an improved elementary motion detector model incorporating a double-layer ON_OFF channel.
• Implemented a cross-type computational architecture for motion detection.
• Designed spatial and temporal consistency detection methods tailored for compound eye structures.

Main Results:

• The proposed model effectively suppresses stationary backgrounds and enhances moving targets.
• Spatial and temporal consistency methods significantly improved detection accuracy and stability.
• Experimental results demonstrate full utilization of image features for moving object detection in complex scenes.

Conclusions:

• The developed algorithm significantly enhances the detection efficiency of moving objects in complex environments.
• The method is well-suited for applications requiring robust motion detection with biomimetic compound eye cameras.
• This advancement addresses key limitations in current motion detection technologies for wide-field imaging systems.