Please respect the original work.Reprint must indicate the link to this article.And the author of the article: slandarer

How to Use MATLAB to Retrieve All Illustrations from Nature (Including Images from 2023-2025)

All image resources are at the end of the article.

Two years ago, I wrote about how to use MATLAB to scrape all illustrations from Nature. Recently, someone asked me if I had a compressed package of the downloaded illustrations for 2024 and 2025, so I ran the code again and found that the code I wrote two years ago still works today. The code is as follows:

function getNaturePNGWhileTure(YEAR)
if nargin < 1
    YEAR = 2024;
end

pbegin = 1; ibegin = 1; jbegin = 1; 
forderName=['Year_',num2str(YEAR)];
if exist(['.\image_',forderName,'\pijbreak.mat'],'file')
    load(['.\image_',forderName,'\pijbreak.mat']);
end
if ~exist(['.\image_',forderName],'dir')
    mkdir(['.\image_',forderName]);
end
disp([pbegin,ibegin,jbegin])

url_full = 'https://www.nature.com/nature/research-articles?searchType=journalSearch&sort=PubDate&year=<Y/>&page=<P/>';
url_year = strrep(url_full,'&lt;Y/&gt;',num2str(YEAR));

options=weboptions('Timeout',inf);
html_year  = webread(strrep(url_year,'&lt;P/&gt;','1'),options);fprintf('1->')
A_page_num = strfind(html_year,'u-visually-hidden'); 
Z_page_num = strfind(html_year,'data-page="next");
page_num   = html_year(A_page_num(find(A_page_num<Z_page_num,1,'last')):Z_page_num);
page_num   = page_num(32:36);
page_num   = str2double(page_num(abs(page_num)<=57&abs(page_num)>=48));

for p = pbegin:page_num
    url_page  = strrep(url_year,'&lt;P/&gt;',num2str(p));
    html_page = webread(url_page,options);fprintf('2\n')
    A_html_artical = strfind(html_page,'itemprop="name headline"');
    Z_html_artical = strfind(html_page,'data-track-action="view article"');

    for i = ibegin:length(Z_html_artical)
        html_artical = html_page(A_html_artical(find(A_html_artical<Z_html_artical(i),1,'last')):Z_html_artical(i));
        A_artical    = strfind(html_artical,'&lt;a href=');
        Z_artical    = strfind(html_artical,'class="c-card__link u-link-inherit"');
        html_artical = html_artical(A_artical(1)+10:Z_artical);
        html_artical = html_artical(1:find(html_artical=='"')-1);

        for j = jbegin:50
            pbegin = p; ibegin = i ; jbegin = j;
            save(['.\image_',forderName,'\pijbreak.mat'],'pbegin','ibegin','jbegin')
            html_png=webread(['https://www.nature.com/',html_artical,'/figures/',num2str(j)]);
            A_png = strfind(html_png,'aria-describedby');
            Z_png = strfind(html_png,'alt="Fig.');
            
            if isempty(Z_png)
                break;
            else
                url_png  = html_png(A_png:Z_png(find(Z_png>A_png,1)));
                url_png  = ['https:',url_png(strfind(url_png,'src="')+5:end-3)];
                url_png  = strrep(url_png,'lw685','full');
                name_png = ['.\image_',forderName,'\',html_artical(10:end),' Fig-',num2str(j)];
                websave(name_png,url_png,options);
            disp(['Downloading Year-',num2str(YEAR),...
                ' Page-',num2str(p),' Artical-',num2str(i),...
                ' Fig-',num2str(j),':',html_artical])
            end
        end
        jbegin = 1;
    end
    ibegin = 1;
end
end

The usage is very simple. For example, to download images from 2024, run the command in the command line window: <span>getNaturePNGWhileTure(2024)</span>:

The code has breakpoints set, allowing you to pause halfway through the download and continue later. The saved file names correspond to the paper numbers. For example, if I want to see the original text of the interesting image below:

We can see that the article number is s41586-023-06728-8. We just need to search:

https://www.nature.com/articles/s41586-023-06728-8

Next, I will showcase some interesting or visually appealing new images, and at the end of the article, I will provide resources for all images from these three years, totaling over ten thousand images.