无法自动执行该过程：您需要做的事情很复杂，并且会从页面的OCR开始，然后识别标题，提取它们，最后从中构建新图像。 />


请记住，您可能会有版权问题ems：报纸通常受版权保护，并且（未经明确许可）您可能无权执行此操作。这可能意味着来自昂贵的律师的恶劣信件或更糟糕的......

There is no way to automate that process: what you would need to do is complicated, and would start with OCR of the page, then identify the headlines, extract them, and finally build a new image from that.

Do bear in mind that you will probably have copyright problems with this: newspapers are generally subject to copyright and (without explicit permission) you probably do not have permission to do this. Which may mean nasty letters from expensive solicitors or worse...