How to Load a Single Image For CNN Classification?

Hello,

I have trained a CNN model to predict three clasess and now I want to check the model's performance based on a new data set. I have 7 new images in each category and 21 images altogether. I can add all the images at once and check the prediction based on the code I have now, but I want to add a single image instead of all 21 images, and see the results.

I would appreciate if anyone can assist me.

The lines of code I am using for now to add new images are as follows:

newImage= imageDatastore(rootfoldertest,'IncludeSubfolders',true);
ds = augmentedImageDatastore(inputSize,newImage,'ColorPreprocessing','gray2rgb');
imageFeatures = activations(trainedNet, ds, featurelayer, 'MiniBatchSize', 32,'OutputAs', 'columns');
label = predict(classifier,imageFeatures, 'ObservationsIn','columns');
sprintf('The loaded image belongs to %s class', label);

I tried using the line of code below instead of the one I am using above but it did not work.

newImage = imread(fullfile('*.jpg'));

Answer 1

[Deepika Ahlawat]

Hi Vinay,

Can you share the error you're getting on trying to access a single image.

To access the single image, you can just provide the location of that image and pass that on to the augmentedImageDatastore function.

newImage = imread('c:\matlab\imagestore\label1\image1.jpg');
ds = augmentedImageDatastore(inputSize,newImage,'ColorPreprocessing','gray2rgb');

[Vinay Chawla]

Here is the error when I use the line of code:

newImage = imread(fullfile('*.jpg'));

ERRORS