Efficient Fine-Grained Object Recognition in High-Resolution Remote Sensing Images From Knowledge Distillation to Filter Grafting

With the development of high-resolution remote sensing images (HR-RSIs) and the escalating demand for intelligent analysis, fine-grained recognition of geospatial objects has become a more practical and challenging task. Although deep learning-based object recognition has achieved superior performance, it is inflexible to be directly utilized to the fine-grained object recognition (FGOR) tasks of HR-RSIs under the limitation of the size of geospatial objects. An efficient fine-grained object recognition method in HR-RSIs from knowledge distillation (KL) to filter grafting is proposed. Specifically, fine-grained object recognition consists of two stages: Stage 1 utilizes oriented region convolutional neural network (oriented R-CNN) to accurately locate and preliminarily classify geospatial objects. At the same time, it serves as a teacher network to guide students' effective learning of fine-grained object recognition; in Stage 2, we design a coarse-to-fine object recognition network (CF-ORNet), as the second teacher network, which realizes fine-grained recognition through feature learning and category correction. After that, we propose a lightweight model from knowledge distillation to filter grafting on two teacher networks to achieve efficient fine-grained object recognition. The experimental results on Vehicle Detection in Aerial Imagery (VEDAI) and HR Ship Collection 2016 (HRSC2016) datasets achieve competitive performance.