Sea Ship Target Detection Method of Remote Sensing Image

Objective: Sea ship target detecting in both military and civilian application has very important practical value. Lots of domestic and foreign experts have carried out extensive research in this field, but contradictions in robustness and the presence of algorithmic complexity make it difficult to meet the needs of embedded board-level applications. So, this paper designed a real time ship target detection algorithm based on Canny operator and two-parameter CFAR algorithm to meet the DSP+FPGA embedded systems. Method: firstly, we use the Canny operator detect edge information to detect rough location of the ship, then use the two-parameter CFAR algorithm identify targets, and thus retaining the advantages of two-meter CFAR algorithm, while reducing computation time. Result: Experimental results indicate that the method in this paper can detect both bright and dark sea ship target of remote sensing image accurately and efficiently. Conclusion: It can satisfy the system requirements of real-time embedded board ship target detection.