Remove Rows/Columns from Binary Image Matrix

I know that when an image is converted to binary, it is turned into an array of 1s and 0s based on the threshold. I want to remove rows and columns based on the code. I would do a set to remove the columns (and rows) before and after the area I want to keep.

A(:, (columns I want to remove))=[].

I want to use this as a way to crop the binary image down. I am concerned in my project that if I crop the my image into partitions prior prior to being binarized, the results are inconsistent. ie, the total spots I may count in the full image is not equal to the sum of all the cropped images. Below is my code for the full project and an example file to work with.

classdef spot_to_heatmap1_exported < matlab.apps.AppBase

    % Properties that correspond to app components
    properties (Access = public)
        UIFigure                 matlab.ui.Figure
        Menu                     matlab.ui.container.Menu
        OpenFileMenu             matlab.ui.container.Menu
        CloseProgramMenu         matlab.ui.container.Menu
        SpotDetectorLabel        matlab.ui.control.Label
        StartButton              matlab.ui.control.Button
        AnalysisParametersPanel  matlab.ui.container.Panel
        RowsSpinner              matlab.ui.control.Spinner
        RowsSpinnerLabel         matlab.ui.control.Label
        ColumnsSpinner           matlab.ui.control.Spinner
        ColumnsSpinnerLabel      matlab.ui.control.Label
        ColorSchemeforHeatMapButtonGroup  matlab.ui.container.ButtonGroup
        GrayButton               matlab.ui.control.RadioButton
        CoolButton               matlab.ui.control.RadioButton
        AutumnButton             matlab.ui.control.RadioButton
        ParulaButton             matlab.ui.control.RadioButton
        SpringButton             matlab.ui.control.RadioButton
        SummerButton             matlab.ui.control.RadioButton
        UIAxes                   matlab.ui.control.UIAxes
    end


    properties (Access = private)
        fullname %for calling file back later


    end


    % Callbacks that handle component events
    methods (Access = private)

        % Menu selected function: OpenFileMenu
        function OpenFileMenuSelected(app, event)
            [filename,pathname] = uigetfile('*.tif','Select the File to Open');
            app.fullname = [pathname,filename];

            movegui(app.UIFigure, 'center')
            hold(app.UIAxes, 'on')
            imshow(app.fullname, 'Parent', app.UIAxes)



        end

        % Menu selected function: CloseProgramMenu
        function CloseProgramMenuSelected(app, event)
            close all
            app.delete
        end

        % Button pushed function: StartButton
        function StartButtonPushed(app, event)
           
            %Image detection------------------------------------------------

            nRows=(app.RowsSpinner.Value);
            nColumns=(app.ColumnsSpinner.Value);

            image=imread(app.fullname);
            [rows, columns, ~]=size(image);
            nV=rows/nRows; %size of each vertical frame for partioning image %%%%%%i swithed, NH, nV if this start looking weird
            nH=columns/nColumns; %size of each horizontal frame for partioning image
           
            A = zeros(nRows, nColumns); %create empty array based on the user size input
            for i=1:nRows %work through each row (top to bottom)
                for k=1:nColumns %within each row, work through the various columns of the image
                    if k*nH==columns
                        continue
                    end
                    if k*nH ~= columns
                        I=imread(app.fullname);
                        %I=imcrop(I, [((k*nH)-nH)  ((i*nV)-nV)   nH   nV]);   %crop naming is [xmin ymin width height]
                        h=fspecial('gaussian',[3 3],1); %gaussian filter on raw image
                        I2=imfilter(I,h, 'replicate', 'conv');                                                          bw=imfill(bw,'holes'); %fill potential holes in spots                       
                        bw=bwareafilt(bw,[20 500], 8); % [range of acceptable sizes], connectivity: adjust these bounds to adjust the size of the particles that are thresholded
                        stats=regionprops('table', bw, 'Circularity');
                        rowidx=(stats.Circularity < .7); %selects all points with circularity less than .6
                        keepSpots=stats(~rowidx, :); %removes all points previously selected by rowidx
                        totalSpots=height(keepSpots); %converts number found spots from table to single value

                        A(i,k)=totalSpots;
                    end
                end
            end



            cdata=A; %convert matrix for use by heatmap code

           
            %color selection and heatmapping-------------------------------

            if app.AutumnButton.Value
                color=autumn;

            elseif  app.ParulaButton.Value
                color=parula;

            elseif app.SpringButton.Value
                color=spring;

            elseif app.SummerButton.Value
                color=summer;

            elseif app.CoolButton.Value
                color=cool;

            elseif app.GrayButton.Value
                color=gray;
            end

            h=heatmap(cdata, 'Colormap', color);





        end
    end

    % Component initialization
    methods (Access = private)

        % Create UIFigure and components
        function createComponents(app)

            % Create UIFigure and hide until all components are created
            app.UIFigure = uifigure('Visible', 'off');
            app.UIFigure.Position = [100 100 640 480];
            app.UIFigure.Name = 'MATLAB App';

            % Create Menu
            app.Menu = uimenu(app.UIFigure);
            app.Menu.Text = 'Menu';

            % Create OpenFileMenu
            app.OpenFileMenu = uimenu(app.Menu);
            app.OpenFileMenu.MenuSelectedFcn = createCallbackFcn(app, @OpenFileMenuSelected, true);
            app.OpenFileMenu.Text = 'Open File';

            % Create CloseProgramMenu
            app.CloseProgramMenu = uimenu(app.Menu);
            app.CloseProgramMenu.MenuSelectedFcn = createCallbackFcn(app, @CloseProgramMenuSelected, true);
            app.CloseProgramMenu.Text = 'Close Program';

            % Create UIAxes
            app.UIAxes = uiaxes(app.UIFigure);
            title(app.UIAxes, 'Selected File')
            zlabel(app.UIAxes, 'Z')
            app.UIAxes.XTickLabel = '';
            app.UIAxes.YTickLabel = '';
            app.UIAxes.TitleHorizontalAlignment = 'left';
            app.UIAxes.FontSize = 16;
            app.UIAxes.Visible = 'off';
            app.UIAxes.Position = [286 32 355 349];

            % Create AnalysisParametersPanel
            app.AnalysisParametersPanel = uipanel(app.UIFigure);
            app.AnalysisParametersPanel.Title = 'Analysis Parameters';
            app.AnalysisParametersPanel.FontWeight = 'bold';
            app.AnalysisParametersPanel.FontSize = 16;
            app.AnalysisParametersPanel.Position = [11 62 264 319];

            % Create ColorSchemeforHeatMapButtonGroup
            app.ColorSchemeforHeatMapButtonGroup = uibuttongroup(app.AnalysisParametersPanel);
            app.ColorSchemeforHeatMapButtonGroup.Title = 'Color Scheme for Heat Map';
            app.ColorSchemeforHeatMapButtonGroup.Position = [33 14 197 168];

            % Create SummerButton
            app.SummerButton = uiradiobutton(app.ColorSchemeforHeatMapButtonGroup);
            app.SummerButton.Text = 'Summer';
            app.SummerButton.Position = [14 102 67 17];

            % Create SpringButton
            app.SpringButton = uiradiobutton(app.ColorSchemeforHeatMapButtonGroup);
            app.SpringButton.Text = 'Spring';
            app.SpringButton.Position = [14 122 65 17];
            app.SpringButton.Value = true;

            % Create ParulaButton
            app.ParulaButton = uiradiobutton(app.ColorSchemeforHeatMapButtonGroup);
            app.ParulaButton.Text = 'Parula';
            app.ParulaButton.Position = [14 58 65 17];

            % Create AutumnButton
            app.AutumnButton = uiradiobutton(app.ColorSchemeforHeatMapButtonGroup);
            app.AutumnButton.Text = 'Autumn';
            app.AutumnButton.Position = [14 80 63 17];

            % Create CoolButton
            app.CoolButton = uiradiobutton(app.ColorSchemeforHeatMapButtonGroup);
            app.CoolButton.Text = 'Cool';
            app.CoolButton.Position = [14 37 47 17];

            % Create GrayButton
            app.GrayButton = uiradiobutton(app.ColorSchemeforHeatMapButtonGroup);
            app.GrayButton.Text = 'Gray';
            app.GrayButton.Position = [14 15 48 22];

            % Create ColumnsSpinnerLabel
            app.ColumnsSpinnerLabel = uilabel(app.AnalysisParametersPanel);
            app.ColumnsSpinnerLabel.HorizontalAlignment = 'right';
            app.ColumnsSpinnerLabel.Position = [46 207 52 22];
            app.ColumnsSpinnerLabel.Text = 'Columns';

            % Create ColumnsSpinner
            app.ColumnsSpinner = uispinner(app.AnalysisParametersPanel);
            app.ColumnsSpinner.Limits = [1 15];
            app.ColumnsSpinner.Position = [113 207 100 22];
            app.ColumnsSpinner.Value = 3;

            % Create RowsSpinnerLabel
            app.RowsSpinnerLabel = uilabel(app.AnalysisParametersPanel);
            app.RowsSpinnerLabel.HorizontalAlignment = 'right';
            app.RowsSpinnerLabel.Position = [62 245 35 22];
            app.RowsSpinnerLabel.Text = 'Rows';

            % Create RowsSpinner
            app.RowsSpinner = uispinner(app.AnalysisParametersPanel);
            app.RowsSpinner.Limits = [1 15];
            app.RowsSpinner.Position = [112 245 100 22];
            app.RowsSpinner.Value = 3;

            % Create StartButton
            app.StartButton = uibutton(app.UIFigure, 'push');
            app.StartButton.ButtonPushedFcn = createCallbackFcn(app, @StartButtonPushed, true);
            app.StartButton.FontSize = 16;
            app.StartButton.FontWeight = 'bold';
            app.StartButton.Position = [92 15 100 28];
            app.StartButton.Text = 'Start';

            % Create SpotDetectorLabel
            app.SpotDetectorLabel = uilabel(app.UIFigure);
            app.SpotDetectorLabel.FontSize = 30;
            app.SpotDetectorLabel.FontWeight = 'bold';
            app.SpotDetectorLabel.FontAngle = 'italic';
            app.SpotDetectorLabel.Position = [206 426 203 39];
            app.SpotDetectorLabel.Text = 'Spot Detector';

            % Show the figure after all components are created
            app.UIFigure.Visible = 'on';
        end
    end

    % App creation and deletion
    methods (Access = public)

        % Construct app
        function app = spot_to_heatmap1_exported

            % Create UIFigure and components
            createComponents(app)

            % Register the app with App Designer
            registerApp(app, app.UIFigure)

            if nargout == 0
                clear app
            end
        end

        % Code that executes before app deletion
        function delete(app)

            % Delete UIFigure when app is deleted
            delete(app.UIFigure)
        end
    end
end

Answer 1

[dpb]

A(:, (columns I want to remove))=[].

will remove everything from the top of the array to the bottom -- if there is a "spot" anywhere in those columns, you'll remove it, indeed, even though you may have only identified one specific location towards the top of the image; when you then do this, there won't be anything left to analyze when your loop over rows gets down to that next set of rows where the spot was.

You can't decimate the image until you've outlined all the areas of interest and, unless the ROIs don't overlap at all, you can't decimate the columns (or rows, either) that are within the overall bounds of all the ROIs.

IOW, you're trying to decimate too soon; you have to do all the ROI identification first and then save those pieces; you can't remove "holes" in the middle of an array (image).

To reiterate, you could only remove from the original image columns/rows that have no intersections with any ROIs across the whole image; how much or little that might be would all depend upon the number, size and relative placement of the ROIs.