Qt-based library with functionality to create highly efficient and fully graphical applications, oriented to computer vision, image processing, and scientific computation. The library features an homogeneous and well documented object-oriented API, with wrapping methods for high performance functionality from libraries such as OpenCV, GSL, CGAL, IPP, BLAS, LAPACK, or Octave library.
Trusion is a physical intrusion system that uses users' Webcams to detect motion, which serves as an indicator to start recording video. Optionally, users can choose to have the program upload the videos to an FTP server in the unfortunate event your computer gets stolen or the perpetrator manages to delete the user's local video files. This is a very new project that is enthusiastically welcoming new contributors and their ideas. This project was build using Qt and OpenCV.
SysopView shows information in a visually attractive way, like computers in movies do. The idea is that it can show information relevant to system administrators non-comprehensible for non-technical personnel, while still looking attractive even if you don't know what is shown. The current version can show nagios status data, webcam streams (mjpeg, jpeg, and video4Linux), and VNC, has a built-in terminal emulator in which regular text applications can be run, can sniff the network (including remotely) and dissect streams and show pictures which are sniffed, can show external bitmaps (from, for example, mrtg and rrdtool), and can scroll text (currently showing output of rssfeeds).
Libcvautomation is a GUI automation and testing tool based on image recognition and response. This program was designed as a direct replacement for Sikuli and Xpresser. Each solution had large problems with crashing, and both refused to function on Red Hat Linux and Ubuntu 12.04. The author really liked the way each of these programs approached GUI automation, but they simply didn't work. As such, a simple library was designed to integrate OpenCV and XTest, which can be used by BASH to drive GUI testing and automation, and works on both new and old Linux distributions.
Darwin Wallet enables blind and visually impaired users to quickly recognize paper money. It uses the phone's camera to detect currency denominations for U.S., U.K., Canadian, European Union, and Australian bills. Simply hold the bill in front of the phone's camera, and the app will speak its value. The app uses the phone's camera and OpenCV (via Android NDK) to analyze and identify the currency image.