hellloxiaotian / ADNet

Attention-guided CNN for image denoising(Neural Networks,2020)
https://doi.org/10.1016/j.neunet.2019.12.024
117 stars 5 forks source link

结果图 #12

Open chenyun-cy opened 4 years ago

chenyun-cy commented 4 years ago

您好,我想问一下,我跑完程序结果只有峰值信噪比以及模型,没有图片的输出,像您论文中图6,图7那种对比图是怎么得到的呢?很希望能得到您的解答,十分感谢!(实在不好意思,刚刚接触pytorch)

hellloxiaotian commented 4 years ago

嗨,最后的输入直接转为python数组,之后重构图像就行~

在 2020-10-13 14:43:03,"chenyun-cy" notifications@github.com 写道:

您好,我想问一下,我跑完程序结果只有峰值信噪比以及模型,没有图片的输出,像您论文中图6,图7那种对比图是怎么得到的呢?很希望能得到您的解答,十分感谢!(实在不好意思,刚刚接触pytorch)

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

chenyun-cy commented 4 years ago

谢谢,我在弄一下

Ijustakid commented 3 years ago

您好,我在test_c.py中想保存source、noisy和out三个图片,我转ndarray后发现有四个维度(3, 3, 702, 1020),其中肯定是[channels, width, height],请问下还有个3是代表啥意思? INoisy = ISource + noise ISource = Variable(ISource) INoisy = Variable(INoisy) ISource = ISource.cuda()

ISource_img = ISource.cpu().numpy() source_path = './result/img_source/' print(ISource_img) print(ISource_img.shape) cv2.imwrite(source_path + 'a.jpg', ISource_img)

》》》(3, 3, 702, 1020)

谢谢

Ijustakid commented 3 years ago

您好,我在test_c.py中想保存source、noisy和out三个图片,我转ndarray后发现有四个维度(3, 3, 702, 1020),其中肯定是[channels, width, height],请问下还有个3是代表啥意思? INoisy = ISource + noise ISource = Variable(ISource) INoisy = Variable(INoisy) ISource = ISource.cuda()

ISource_img = ISource.cpu().numpy() source_path = './result/img_source/' print(ISource_img) print(ISource_img.shape) cv2.imwrite(source_path + 'a.jpg', ISource_img)

》》》(3, 3, 702, 1020)

谢谢

搞定了,感谢

Ruobo-Xu commented 3 years ago

您好,我在test_c.py中想保存source、noisy和out三个图片,我转ndarray后发现有四个维度(3, 3, 702, 1020),其中肯定是[channels, width, height],请问下还有个3是代表啥意思? INoisy = ISource + noise ISource = Variable(ISource) INoisy = Variable(INoisy) ISource = ISource.cuda() ISource_img = ISource.cpu().numpy() source_path = './result/img_source/' print(ISource_img) print(ISource_img.shape) cv2.imwrite(source_path + 'a.jpg', ISource_img) 》》》(3, 3, 702, 1020) 谢谢

搞定了,感谢

您好,可以给我一份吗,我想把生成的图保存下来,万分感谢

Ruobo-Xu commented 3 years ago

您好,我在test_c.py中想保存source、noisy和out三个图片,我转ndarray后发现有四个维度(3, 3, 702, 1020),其中肯定是[channels, width, height],请问下还有个3是代表啥意思? INoisy = ISource + noise ISource = Variable(ISource) INoisy = Variable(INoisy) ISource = ISource.cuda() ISource_img = ISource.cpu().numpy() source_path = './result/img_source/' print(ISource_img) print(ISource_img.shape) cv2.imwrite(source_path + 'a.jpg', ISource_img) 》》》(3, 3, 702, 1020) 谢谢

搞定了,感谢

您好,我在test_c.py中想保存source、noisy和out三个图片,我转ndarray后发现有四个维度(3, 3, 702, 1020),其中肯定是[channels, width, height],请问下还有个3是代表啥意思? INoisy = ISource + noise ISource = Variable(ISource) INoisy = Variable(INoisy) ISource = ISource.cuda()

ISource_img = ISource.cpu().numpy() source_path = './result/img_source/' print(ISource_img) print(ISource_img.shape) cv2.imwrite(source_path + 'a.jpg', ISource_img)

》》》(3, 3, 702, 1020)

谢谢

您好,可以给我一份吗,我想把生成的图保存下来,万分感谢

Ruobo-Xu commented 3 years ago

谢谢,我在弄一下

您好,保存下来了吗

hellloxiaotian commented 3 years ago

您好!非常感谢您的关注!原来只保存了ADNet论文中展示可视化结果,其余结果没有保存。此外,博士刚毕业,还没有去新单位入职。因此,不能用服务器测试模型,构造可视化图像。

祝好~ 春伟

在 2021-01-26 20:49:26,"Ruobo-Xu" notifications@github.com 写道:

谢谢,我在弄一下

您好,保存下来了吗

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

thestars-maker commented 11 months ago

请问如何保存去噪之后的图片啊,有大佬可以分享一下吗

hellloxiaotian commented 10 months ago

您好:

和正常的python保存图像一样,可以参考百度上一下python如何保存图像的。

祝好 田春伟 副教授 西北工业大学

在 2023-12-08 17:00:42,"thestars-maker" @.***> 写道:

请问如何保存去噪之后的图片啊,有大佬可以分享一下吗

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>