13,000 Screen Capture Images + How to Get More

在其站点上为图像分配链接的方式使扫描和从站点上刮取图像变得容易。URL代码是根据长度为6的字母和数字的组合顺序生成的。 将代表一个接一个上传的图像。当然，编写脚本来刮取和收集图像非常容易。到目前为止，我总共收集了1,000,000张图像，而我现在正在这里制作尽可能多的图像，但是您当然可以想象从该站点收集大量图像非常容易。我的图像是在过去2年中收集的，尽管绝大多数图像是在最近4个月内收集的。The way links are assigned to images on this site enables convenient scanning and scraping of images from the platform. URL codes are generated using sequential combinations of letters and digits with a fixed length of 6. Each such URL corresponds to images uploaded in sequence. Naturally, it is highly feasible to write scripts for scraping and collecting these images. To date, I have accumulated a total of 1,000,000 collected images, and I am still gathering as many as possible here at present. Needless to say, collecting large volumes of images from this site is remarkably straightforward, as anyone can readily imagine. The images in my dataset have been gathered over the past two years, with the vast majority being acquired within the last four months.

搜集汇总

背景与挑战

背景概述

该数据集包含13,000张从prnt.sc网站抓取的屏幕截图图像，并附有如何收集更多图像的说明。这些图像通过顺序URL代码易于抓取，适用于图像分类和机器学习实验，内容涵盖游戏、人物、文本等多种类别。