Change detection of large-field-of-view video images in low-light environments with cross-scale feature fusion and pseudo-change mitigation

Key Points

• Change detection algorithms enhance accuracy of identifying real changes in video images, especially in low-light settings.
• The method employs cross-scale feature fusion to effectively combine data from various scales for improved detection accuracy.
• The approach addresses challenges posed by low-light environments and minimizes false positives through pseudo-change mitigation techniques.
• The implementation highlights the need for further validation in varied lighting conditions and across different environments.