基于FPGA的红外焦平面成像条纹噪声逐帧抑制算法的研究

Strip Noise Suppression Algorithm of Infrared Focal Plane Imaging Frame by Frame Based on FPGA

红外焦平面阵列是当今红外成像技术发展的主要方向,随着器件工艺的进步,红外焦平面阵列探测器有了长足的发展,然而红外图像普遍具有信噪比低的缺点,这大大限制了红外焦平面的应用。与固定图案噪声(FPN)相比,随机噪声的最大特点是每帧均不同,因此去除该类型噪声的算法必须在一帧之内完成。提出了一种新型的单帧去除此类条纹噪声的算法并加以硬件实现,在单帧内设置适合的校正参数和阈值,达到在单帧图像内有效去除条纹噪声的目的。通过算法处理前后的图像对比以及客观的MSE、PSNR测试数据对比,证明了该算法能够有效地改善焦平面器件成像质量。

IRFPA is the main development direction in the area of infrared imaging technology. In recent years, with the de- velopment of the device technology, infrared focal plane array detector is improved successfully. The application of IRFPA is limited, because of the infrared image with low SNR. As is well known to us, the biggest difference between the random noise and the FPN is that the strip noise＇s features of each frame are different. So, the new method to eliminate this kind of noise must be done in one frame. And we need to set adaptive corrective parameters and threshold values. A new method is put forward in this paper and the hardware implementation has been done at the same time. Both the subjective images and objective data MSE.PSNR indicate that the new method raised in this paper can improve the image quality effectively.