摘要
针对舰船边缘信息检测实时性差、检测效果不佳等问题,通过选取适用于舰船边缘检测的数学形态运算与最大类间方差二值法的方法,并结合FPGA并行处理能力,构建舰船边缘检测硬件平台。在该硬件平台上,采用FPGA对舰船目标相关算法进行硬件实现。从舰船图像处理结果分析可知,采用传统CPU进行算法处理需要约120ms,采用FPGA实现仅需21.2ms,采用FPGA加速后处理速度提高5倍以上,完全可满足实时性要求,因此该方法具有良好的应用前景。
In order to solve the problems of poor real-time performance and poor detection effect in ship edge information detection, this paper constructs a hardware platform for ship edge detection by choosing the methods of mathematical morphology operation and OSTU,and combining the parallel processing ability of FPGA. On this hardware platform,we use FPGA to implement the related algo-rithm of ship target. From our analysis of the results of ship image processing,it takes about 120ms to process the ship image using tra-ditional CPU. It takes about 21.2ms to implement by using FPGA. The speed of post-processing accelerated by using FPGA is five times faster than traditional CPU,which can meet the real-time requirements and has broad application prospects.
作者
李世文
李立聪
LI Shi-wen;LI Li-cong(Heyuan Polytechnic College,Heyuan 517000,China;Shenzhen Jiamusi Technology Co.,Ltd.,Shenzhen 518110,China)
出处
《软件导刊》
2019年第6期176-180,共5页
Software Guide
基金
广东省高等职业教育品牌专业建设项目(粤教高函(2016)114)
广东省河源市市级项目(170608090221295)
作者简介
李世文(1979-),男,硕士,河源职业技术学院高级工程师,研究方向为电子设计、图像处理;李立聪(1983-),男,深圳佳姆斯科技有限公司技术工程师,研究方向为电子电路设计.
