Brain-inspired Visual Perception of Low-altitude Slow and Small Targets：Current Research and Future Perspectives

Low-altitude Slow and Small （LSS） targets pose significant detection challenges for conventional radar systems due to their low flight altitude，slow velocity，small radar cross-section，and ability to hide within complex backgrounds （e.g.，urban structures，mountainous terrain，or forests）. Photoelectric sensors （including infrared and visible spectrum） coupled with intelligent visual perception technologies，particularly deep learning，offer promising solutions to these limitations. However，the diversity of LSS target signatures，the complexity of detection environments，and the openness of application scenarios collectively constrain the transferability and generalization capability of trained deep learning models. Current research focuses on drawing inspiration from the Human Visual System （HVS），specifically its capacity for integrated appearance-motion analysis during visual information acquisition，processing，and interpretation. This biomimetic approach has emerged as a key direction for advancing LSS target perception. This paper systematically reviews the state-of-the-art technological developments while providing critical insights into future research in this field.