An m-by-n-by-1 image cannot be used as input image in the Fully Convolutional Network FCN ?

1 次查看（过去 30 天）

Gobert 2021-12-7

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1605335-an-m-by-n-by-1-image-cannot-be-used-as-input-image-in-the-fully-convolutional-network-fcn

评论： Matt J 2021-12-9

Hi, Can you help? I am trying to use the FCN (fully convolutional network) layers for semantic segmentation. Here's the function I used:

lgraph = fcnLayers(InputImageSize, NumberOfClasses);
 net = trainNetwork(dstrain,lgraph,options);

Here's the error I got:

Error using trainNetwork (line 184) The training images are of size 256×256×1 but the input layer expects images of size 256×256×3. Error in network_bx (line 114)

My question: I would like to know if FCN layer does not work on m×n×1 or grayscale images. If it does, can you help me to understand why I got the above error when I used images with the size m×m×1?

2 个评论
显示无隐藏无

Matt J 2021-12-8

What did you give as InputImageSize?

Matt J 2021-12-8

Robert replied:

I used images of input size 256 × 256 × 1

请先登录，再进行评论。

请先登录，再回答此问题。

回答（2 个）

Matt J 2021-12-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1605335-an-m-by-n-by-1-image-cannot-be-used-as-input-image-in-the-fully-convolutional-network-fcn#answer_850110

I was able to modify the input size in deepNetworkDesigner. No idea what will happen when you try to train it.

14 个评论
显示 12更早的评论隐藏 12更早的评论

Gobert 2021-12-9

Did you successfully train the fcn after doing your modifications?

Matt J 2021-12-9

I did not attempt to train. My GPU isn't that great.

请先登录，再进行评论。

yanqi liu 2021-12-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1605335-an-m-by-n-by-1-image-cannot-be-used-as-input-image-in-the-fully-convolutional-network-fcn#answer_850325

yes，sir，may be change the data load，such as

imageSize = [256 256 3];

augimds = augmentedImageDatastore(imageSize,dstrain,'ColorPreprocessing','gray2rgb');

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

yanqi liu 2021-12-9

在 MATLAB Online 中打开

yes，sir，may be use

trainingImages = imageDatastore('train',...
    'IncludeSubfolders',true,...
    'LabelSource','foldernames','ReadFcn',@data_preporcess);
	
function data = data_preporcess(file)
data = imread(file);
if ndims(data) == 2
    data = cat(3, data, data, data);
end
data = imresize(data, [256 256], 'bilinear');
data = double(data);

请先登录，再进行评论。

请先登录，再回答此问题。

类别

AI and Statistics Deep Learning Toolbox Image Data Workflows

在 Help Center 和 File Exchange 中查找有关 Image Data Workflows 的更多信息

产品

MATLAB

版本

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by