Having the audacity to databend images

Whilst I was researching about how to intentionally glitch digital images, I came across the term ‘databending’. Basically, this means opening a file in a piece of software which it wasn’t meant to be used for. For example, using Audacity, which is a sound editor, to open an image file, also known as misalignment.

First of all, I needed to convert my chosen image to a TIFF file, so that there was more data to play around with. After importing the image in Audacty as raw data, the file is converted into an audio wave form. Then I selected a section of the ‘audio’ to add an effect, such as echo and reverb. This then alters the data, which ultimately will distort and glitch the original image. 

Below are some of my experiments with data bending an image with Audacity.  

Echo and reverb

Echo and reverb

Wave form for the image above.

Wave form for the image above.

Fade in and echo

Fade in and echo

Whilst in Audacity, it is impossible to know what the resulting image will look like, until it is exported and opened. This reminded me of the times when I would open my pack of photographs from Boots to see if the negatives from my camera film had developed corrrectly. Sometimes, the photos would be damaged by being double exposed or negatives overlapping. 

For my final assignment, I am going to explore the different audio effects further, using one image.  


Screen Shot 2018-09-22 at 17.12.34.png
Screen Shot 2018-09-22 at 17.11.43.png
Screen Shot 2018-09-22 at 17.12.07.png
Screen Shot 2018-09-22 at 17.10.59.png

I’ve have started to develop my ideas from Assignment 4, to explore how I can create pixelated images without filters. The images above have been produced from enlarging a photograph to 800% and then taking screen shots of areas which I think are of particular interest. I am wondering whether there is a way that I can taken sections of an enlarged image and then combining them to resemble the original.

Screen Shot 2018-09-22 at 17.36.25.png
Screen Shot 2018-09-22 at 17.34.44.png
Screen Shot 2018-09-22 at 17.32.52.png
Screen Shot 2018-09-22 at 17.29.10.png

Jesse wants a new haircut.


At first glance, it is difficult to know which of the two images of Jesse Lingard is a photograph. Computer-generated images are so realistic nowadays, it is not surprising that this type of problem has arisen. If images can be created so accurately, they may come a time when cameras are no longer needed. Graphics are so realistic that there might not be a need in the future to send a photographer out, when the ideal image can be made on screen. 

However, it does raise the issue of image rights when creating a representation of a person. Also, what are the responsibilities of the image creator once it is out in the public domain.

At one level, it could be argued that the collection of pixels are not real, and would not exist without the screen. However, by being placed within a game-based scenario, the character becomes life-like. As computer generated imagery becomes more prevalent, it will be interesting to see what responsibilities will need to be granted for creators and what rights there are for the sprites.  


In an earlier post, I mentioned how Quick Response (QR) codes appear to be not used to their full potential, especially in terms of photography. 

A QR code is a small square, made up of black and white squares. Similarly to a bar code, the. QR code can be scanned and link to other relevant information.  

If the QR code was part of a photograph, then it could be perceived as an invisible layer. I like the potential it could have for enabling the viewer to interact with the subject at s deeper level. This could include revealing a map of the photograph's location, or an audio clip that would provide a narrative. 

Despite these apparent advantages, having a black and white square on the photograph would obstruct the viewer from appreciating the image in its entirety. Ideally, I would want the QR code to be hidden.