Date of Award
Master of Science
Dr. John Moya, and associated research assistants, have previously created an image-change recognition algorithm (JESSE) to mark changes within an image. The focus of this thesis is to present a physical application and modification of this algorithm in order to detect a surgeon's hand and verify chip placement on a printed circuit board.
There are different techniques in implementing visual recognition and motion detection with smart systems but the high cost and complicated calibration of these systems make them impractical. The goal was to create a system that is simple, inexpensive and applicable to multiple applications that will allow the user maximum flexibility.
The hardware necessary to provide input to the JESSE algorithm consists of a CMOS sensor camera, a light platform that serves as a stage and a laptop/pc with the necessary software (MATLAB and Visual Studio C# 2010). The Platform itself includes an adjustable stand for the CMOS camera, as well as an optional LCD, 16x16 LED Dot Matrix, pan and tilt servos, and a power supply. An algorithm that runs in the background as a user friendly GUI provides feedback on object targeting and was made in Visual Studio with cross-platform implementation to run on 32 and 64 bit architectures.
The JESSE algorithm itself samples two images, a reference and a test, via four Gaussianweighted filters. Then using three of the Gaussians filters' results triangulates the position of change in the image. The location of this change may be marked in a new image, creating an output image with reference crosshairs.
The hardware as well as software platform is able to detect movement of a surgeon's hand or the absence of chip on a pc board, thus providing x and y change coordinates in a 2D space. The resulting coordinates x, y can be used for multiple applications such as a virtual mouse, running the pan and tilt on a second camera using two servos, or plotting to a dot matrix display, all of which have been implemented using the platform created from this thesis. There are also further possibilities for utilizing the resulting coordinates, as will be discussed at the end of this thesis.
Received from ProQuest
Miguel Angel Chaidez
Chaidez, Miguel Angel, "2D Gaussian Object Motion Detection" (2011). Open Access Theses & Dissertations. 2257.