Survey Of Small Object Detection Methods Based On Deep Learning

With the rapid development of computer vision and artificial intelligence technology, object detection has received more extensive attention. Due to the small proportion of small object pixels, the lack of semantic information, the susceptibility to complex scene interference, and the easy aggregation and occlusion, small object detection has always been a major difficulty in the field of object detection. At present, small object detection of vision is becoming increasingly important in all areas of life. Through a lot of research and learning, this paper introduces the current difficulties encountered in small object detection and systematically summarizes the small object detection methods from the aspects of multi-scale feature fusion, anchor frame mechanism, optimized IoU threshold, contextual semantic information and attention mechanism.