Aircraft detection from high-resolution remote sensing images is important for civil and military applications. Recently, detection\nmethods based on deep learning have rapidly advanced. However, they require numerous samples to train the detectionmodel and\ncannot be directly used to efficiently handle large-area remote sensing images. A weakly supervised learning method (WSLM) can\ndetect a target with few samples. However, it cannot extract an adequate number of features, and the detection accuracy requires\nimprovement.We propose a cascade convolutional neural network (CCNN) framework based on transfer-learning and geometric\nfeature constraints (GFC) for aircraft detection. It achieves high accuracy and efficient detection with relatively few samples. A\nhigh-accuracy detection model is first obtained using transfer-learning to fine-tune pretrained models with few samples. Then, a\nGFC region proposal filtering method improves detection efficiency. The CCNN framework completes the aircraft detection for\nlarge-area remote sensing images. The framework first-level network is an image classifier, which filters the entire image, excluding\nmost areaswith no aircraft.Thesecond-level network is an object detector,which rapidly detects aircraft fromthe first-level network\noutput. Compared with WSLM, detection accuracy increased by 3.66%, false detection decreased by 64%, and missed detection\ndecreased by 23.1%.
Loading....