Saturday, April 27, 2013
Search for Privacy Leaks through Images
We live in a society that depends heavily on the use of technology, specifically the Internet, to complete our daily activities and to stay in touch with friends and family. Perhaps one of the most potentially dangerous activities on the Internet is keeping in touch with others via social media, or social networking sites. This exploratory study examined personally identifiable information (PII), of photos uploaded to Online Social Networks (OSN) like Twitter, facebook etc.
We collected images from various online social media . user upload the images in these social medias using web services. some of the web services extract the exif data of the image but most does not. We collect the images which has meta data embedded in it. We extracted the meta data of the imege. Which includes the device information and geo location of the image. This geo-locations helps us locating the persons visited location and where we can find him.
We collected images which are being uploaded by web services which does not extract the metadata of the image.
EXIF DATA : Exchangeable information .this is the metadata of the image which contains all the information of the image. Like the image is blurred or not, tampered or not, geo graphical co –ordinates of the image , device from which image has been taken. Our study mainly focuses on the geographical co ordinates of the image.
GEO –INFORMATION : the meta data of the image gives the exact latitude and longitude of a place. By placing these co ordinates in Google map we can trace where the person last visited .
People click their images in the high end devices when they visit some place and upload them in the social networks. Which is actually a privacy threat and the user is unknown to this fact. We collected around 500 images from different photo uploading web services like twit goo flicker,twitpic,yfrog,twitcli. Put these images into the exif data code .that extracts all the raw data of the image . some of the web services used to upload data extracts metadata of the image and some web services does not. We concentrated on the web services which does not extract metadata. the meta data extractor code is written in java language. Then we analysed the raw data we are getting from the image meta data. our focus is on geo graphical co ordinates of the image. From the images we collected we could able to find what the places the person visited are.