Removing gray border from imshow plot

I am trying to create images from 128x128 matrices of complex doubles and then use those images as my test data for a convolutional neural network using the Deep Learning Toolbox(my training images are 128x128 .png files). I am using the imshow command to plot the data from the matrices because that command removes the axes and tick markings from the plot, but this command is leaving an unwanted gray border around my images. Instead of getting the desired 128x128 images, I'm getting images that are 185x160. I also tried using the imsave command, but the saved image turns out to just be a black box and does not match the plotted figure. Is there any other way for me to get only the desired 128x128 images without any borders or axes? Here is my current code:

figure, imshow( complex_image_new_dB, image_scale_bounds_new_c );
axis equal;
set(gca,'units','pixels'); % set the axes units to pixels
x = get(gca,'position'); % get the position of the axes
set(gcf,'units','pixels'); % set the figure units to pixels
y = get(gcf,'position'); % get the figure position
set(gcf,'position',[y(1) y(2) x(3) x(4)]);% set the position of the figure to the length and width of the axes
set(gca,'units','normalized','position',[0 0 1 1]); % set the axes units to pixels

Answer 1

[Chunru]

try the following command

imwrite(X,MAP,FILENAME,FMT) 

where X is the data matrix you have (it should be real), MAP is the colourmap you use, FMT should be 'png'.

[DGM]

Absolutely correct. If the image data is available, just use imwrite. There's one additional thing though. OP's image data is likely not scaled to standard image data range, hence the syntax used for imshow() and the issues with black output.

% assuming images are grayscale
imwrite(mat2gray(myimage,mydatarange),'unnecessarilylongfilename.png')

mat2gray() is going to rescale the image to standard data range based on the specified range and the numeric class of the image. If the range is left unspecified, it uses [min max] automatically.