为了从单片Bayer格式图像探测器中获取更高分辨率实时彩色图像, 设计了一套基于FPGA硬件的彩色图像实时采集系统。重点研究了影响彩色图像质量的Bayer格式图像去马赛克放大算法, 该算法首先估计原始Bayer格式图像中丢失的绿色通道信息; 然后计算颜色差R-G和B-G; 接着将梯度方法用于确定丢失的绿色像素, 放大绿色通道; 最后, 放大颜色差值图像R-G和B-G, 再结合放大的绿色通道, 便可恢复出红色和蓝色通道信息, 得到放大全彩色图像。为了减少数据量, 设计了基于FPGA的JPEG压缩功能, 同时设计了千兆以太网(GigE)图像传输机制, 使系统数据吞吐率大大提高。试验数据表明, 在分辨率为2352×1728、帧频为20f/s的情况下, 可以实时输出全彩色放大图像。目前, 系统已经工程应用, 其优异表现满足了项目实际需要。

In order to obtain real-time zoomed full color image with a single image sensor coated with Bayer CFA, proposed is a real-time demosaicking and zooming image acquisition system based on FPGA. The demosaicking and zooming algorithm which is critical to the color image quality was studied in detail. For this algorithm, it is initially to calculate the missing G, then to compute the two color difference images R-G and B-G so as to reduce the demosaicking and zooming errors. Next, it is to choose the missing zoomed G by calculating and comparing the chrominance gradient. Finally, the enlarged red and blue channels are obtained by adding the corresponding enlarged R-G and B-G color difference images to the enlarged green channel. In order to reduce the data amount, FPGA-based JPEG was designed and GigE based image transmission was implemented to improve the image throughput. Experimental results prove that real-time 2352×1728 resolution demosaicking and zooming image can be achieved at the speed of 20f/s.