Search

The sophistication of ship detection technology in remote sensing images is insufficient, the detection results differ substantially from the practical requirements, mainly reflected in the inadequate support for the differentiated application of multi-scene, multi-resolution and multi-type target ships. To overcome these challenges, a ship detection method based on multiscale feature extraction and lightweight CNN is proposed. Firstly, the candidate-region extraction method, based on a multiscale model, can cover the potential targets under different backgrounds accurately. Secondly, the multiple feature fusion method is employed to achieve ship classification, in which, Fourier global spectrum features are applied to discriminate between targets and simple interference, and the targets in complex interference scenarios are further distinguished by using lightweight CNN. Thirdly, the cascade classifier training algorithm and an improved non-maximum suppression method are used to minimise the classification error rate and maximise generalisation, which can achieve final-target confirmation. Experimental results validate our method, showing that it significantly outperforms the available alternatives, reducing the model size by up to 2.17 times while improving detection performance be improved by up to 5.5% in multi-interference scenarios. Furthermore, the robustness ability was verified by three indicators, among which the F-measure score and true-false-positive rate can increase by up to 5.8% and 4.7% respectively, while the mean error rate can decrease by up to 38.2%.

With the continuous enhancement of remote sensing technology, using satellite to detect and identify targets has important research significance in military and civil fields. Due to the influence of many interference factors, the background of large-scale infrared remote sensing images is often complex and the target intensity is weak, and efficient and accurate target detection has become a difficult issue. Therefore, the purpose of this paper is to solve the issues of inaccurate target detection and inefficient hardware deployment. Firstly, the rough detection method based on the grey variance and gradient features can effectively extracted the target candidate region and eliminate simple background interference. Secondly, the fine detection method based on multiple features is used for ship classification, in which structured LBP (local binary pattern) features is applied to calculate the structured feature histogram, screen the local target features, and calculate and classify the global features through two-dimensional Fourier features. Thirdly, the improved non-maximum suppression method is used to merge adjacent target points to ensure the accuracy of the algorithm to the greatest extent. Finally, an efficient feature computing unit and multichannel storage structure is proposed to optimize the FPGA (field-programmable gate array) implementation and relieve the pressure of high data rate processing and can eliminate a lot of computational redundancy. Experimental results validate that, compared with some advanced methods, our method can reduce the processing time by up to 1.71 times while the ship detection performance can be improved by up to 5.4%. Furthermore, the hardware processing capability was verified by three indicators, among which the traversal processing performance and throughput can increase by up to 1.7 and 6.7%, while the delays of key modules can decrease by up to 4%. [ABSTRACT FROM AUTHOR]