This happens a lot on Twitter. Someone posts an archival photo that would make a great visual for a lecture. But tweeted images are not of sufficient quality to be effective in a Powerpoint presentation and not everyone includes the source in their tweets. For instance, @
There are a few easy options for tracking the source of an image to find a version with a higher resolution. Google Images offers image searches, which allow users to upload an image from their computer, or just copy and paste the url of an image, and it will seek out similar versions from across the internet. Sure enough, it brought me to the website of the National Fairgrounds Archive at the University of Sheffield. Google also offers a selection of ‘visually similar images,’ but the results bring up everything from a slave market in the Caribbean to a schoolhouse in the Solomon Islands.
Another tool that can help find a better version of an image is Tineye. This app was developed by Idée Inc. as a ‘reverse image search.’ Tineye works much the same as Google, and for matching an exact image the results are probably the same. One of the benefits of Tineye is that it will rank results according to ‘closest match,’ ‘most changed,’ and ‘biggest image.’ The first and last options are useful if you are looking for the most accurate version or for an image large enough to be included on a Powerpoint slide. The ‘most changed’ option is where Tineye really stands out. This option makes Tineye much better at finding altered versions of an image. I decided to experiment with one of the more famous altered images of the First World War, a composite image by Frank Hurley.
This image is one of three similar compositions that placed a photograph of a dressing station under a much more dramatic sky.
Pasting the url for the composite image into Tineye and searching for ‘most changed’ brought up a few different versions of the image, with variations in contrast and shading, the cover of Somme Mud, which featured the photo on the frontpiece, as well as a few hits of the original image with a link back to the Australian War Memorial website. Tineye was able to de-compose the image, find the original photograph, and identify its source.
I should not oversell Tineye’s capabilities. It was able to decompose one photograph, but I tried it with compositions that mixed three or more images together with little success. At best, Tineye was able to find different versions of the same composite image, but not the original photographs. While it cannot decompose all composites, a reverse image search that shows how images have been altered over time holds some promise for a historian. At the very least, Tineye and Google Images will help me find a photograph large enough to include in my lecture slides.