Aponte-Medina, David

Profile Picture

Filters

20101
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Publication
    Computer vision system for translucent medical devices monitoring during manufacturing process
    (2010) Aponte-Medina, David; Torres-Muñiz, Raúl E.; College of Engineering; Juan-García, Eduardo J.; Toledo-Quiñones, Manuel; Department of Electrical and Computer Engineering; Colón-Rivera, Celia
    Computer vision is one of the most important technologies for monitoring and control used in the manufacturing industries and is fast emerging in recent applications. The processes that include automatic computer vision techniques are generally faster, more reliable, deterministic, and productive. These systems avoid some human limitations like fatigue and errors. This project, investigates one important part of the automation strategy to enable the automatic assembly and test of deformable life critical medical device such as catheters. The project is focused on developing and implements a visual inspection computer system and the algorithms necessary to monitor the catheter device unit during the assembly process. The device unit is dispensed and placed into the "initial stage" during a previous manufacturing step. A robot picked up the device unit and transports it to the next manufacturing stage for additional processing. The visual inspection computer system is designed to monitors the robot movement through a controlled predefined path and reacting when the device unit falls from the robot arm into the controlled environment. The designed vision system was able to detect and acquire an image when the device under test falls. Then, it passes this image through the "skeleton track", which is the designed image processing algorithm to identify the coordinates of the extremes (end points) at the image plane. An Associative Neural Network system was used to estimates the coordinates of the device unit extremes at the robot 4D plane based on the image coordinates plane. Finally, the robot coordinates values are transmitted to the robot which move the gripper until it is placed on the appropriate location at the corresponding device unit extremes.