摘要
根据非制冷型红外探测器的成像原理,分析其红外图像的直方图特点,并结合经典红外图像增强算法的优缺点,提出一种适合在FPGA上实时实现的图像增强算法。文中通过分析自适应平台值直方图增强算法增强红外图像细节的原理,以及其相对于经典直方图增强的优势,同时指出其迭代算法在硬件实现上的局限性,提出简化后的平台值直方图增强算法。然后再结合基于背景中值的灰度映射,通过加权平均得到最后图像。经过理论分析和实验证明,本文算法可以很好地移植到FPGA上实现,满足系统实时处理的要求,算法增强后的图像边缘清晰,目标突出。
This paper introduced the adaptive contrast enhancement algorithm based on histogram equalization and analyzed its working principle. According to the advantage it has among classic enhancement algorithm and the limitation of the iterative algorithm's implement on hardware,we proposed a simplified real-time algorithm for infrared image enhancement which is easy to accomplish on FPGA. Combining the algorithm mentioned above with direct gray-level mapping based on the gray of the background,the finial enhanced image is generated by calculating the weighted sum of the two algorithms. Theoretical analysis and experiment results both show that,the proposed algorithm works well and also meets the real-time requirement when transplanted into FPGA.
出处
《电子设计工程》
2017年第24期170-175,共6页
Electronic Design Engineering
