Gangs are a serious threat to public safety throughout the United States, and are responsible for an increasing percentage of criminal and violent acts in many communities. According to US law enforcement officials, gangs commit as much as 80% of the crime in many communities.
Street gang graffiti is the most common way that gangs communicate messages, including challenges, warnings, or intimidation to their rivals. However, such graffiti is an excellent way to track gang affiliation and growth, or even sometimes to obtain membership information.
In fact, that is the aim of the GARI (Gang Graffiti Automatic Recognition and Interpretation) project at Purdue University (West Lafayette, IN, USA) where researchers are developing a system capable of analyzing the hidden messages and identifying characteristics in the images of graffiti.
The image analysis process involves capturing the metadata (geoposition, date, and time) from the image of the graffiti on a mobile device such as a cell phone and the extraction of relevant features such as color and shape from it.
The data are sent to a server and compared against a database that contains other images of graffiti. Matched results are then sent back to the mobile device where the user can review them and provide extra input to further refine the information. Once the graffiti is decoded and interpreted, it is labeled and added back onto the database.
The researchers recently presented a prototype of the system to homeland security officials and law enforcement personnel in Indianapolis.
-- Posted by Vision Systems Design
