基于DeblurGAN对运动图像的去模糊化研究

Deblurring of Moving Images Based on DeblurGAN

在实际生产与生活中,拍摄设备与成像物体之间经常会出现难以保持相对静止的状态,故拍摄出来的图像易出现运动模糊现象。针对该问题,基于DeblurGAN对运动模糊图像进行去模糊化处理。实验结果表明,该模型可以有效去除运动图像中的模糊,同时对比传统算法,该算法不仅能使获得的图像更清晰,同时还能增加图像中的纹理和细节。

In actual production and life,it is often difficult to stay relatively still between the shooting device and the imaging object,so motion blur is easy to appear in the image taken.For this problem,we can use DeblurGAN to deblur motion blur images.The model can effectively re move blur in moving images in the experimental process.At the same time,compared with the traditional algorithm,this algorithm can not only make the obtained image clearer,but also increase the texture and detail in the image.