Is it possible remove a data point when the logical index removes a coordinate?

Hello,

I'm trying to create a scatter plot with coordinates for fixations from an excel table. Trying to create a region of interest that's a certain number of pixels big by logically removing the coordinates outside of the main area. I got a .

%step 3:filter verified saccades for only those that appear in the target areas
rawdata = readtable('RES_IAS (1).xlsx');
fix_data = rawdata(:,3:4);                  %extract the fixation coordinates from raw data
fix_x = table2array(fix_data(1:end,1));     %extracted X-coordinates to array
fix_y = table2array(fix_data(1:end,2));     %extracted y-coordinates to array
idx2 = fix_x > 0; fix_x < 925               %logical index for x
idx3 = fix_y > 0; fix_y < 925               %logical index for y
result1 = [fix_x(idx2) fix_y(idx3)];        %combined x and y coordinates

The code below outputs a scatterplot, I'd like to increase the filter to include a sort of border all the way around the edges (about 150 pixels)

%figure 3: fixations within ROI 
x1 = fix_x(idx2)               
y1 = fix_y(idx3)
figure;
scatter(x1,y1)
xlabel('X coordinates of fixations (px)'), ylabel('Y coordinates of fixations(px)'), title('Areas of fixation')

I'm aware that it's a bit redundant to have 2 indexes of the same value, but if I try to combine them into a single index like

idx2= fix_x, fix_y > 0 

outputs "error:Array indices must be positive integers or logical values".

The index currently only filters a single data point at 0,0. I want to increase the size of the index, say, 95 pixels minimum, but while it removes the Y coordinates, the X-coordinates aren't removed and then I get

"Error using horzcat. Dimensions of arrays being concatenated are not consistent."

Is there a way to get both sets of values removed from the arrays? Or will it need to be edited directly from the table?

Answer 1

[Matt J]

region=x1>0 & x1<925 & y1>0 & y1<925;

scatter( x1(region),y1(region) )

[Ruth Ronalter]

Thanks, I didn't think to make the 'and' an &. Actually was able to edit out about 4 variables (I have a lot for this project), and made the filter closer to the size of my actual regions of interest