#TITLE_ALTERNATIVE#
Object tracking is a system that has an ability to identify and follow the movement of the object being observed. Identification of objects can be distinguished from form, <br /> <br /> <br /> color, texture and other characteristics that can differentiate objects with the surrou...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/15984 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Object tracking is a system that has an ability to identify and follow the movement of the object being observed. Identification of objects can be distinguished from form, <br />
<br />
<br />
color, texture and other characteristics that can differentiate objects with the surrounding environments. <br />
<br />
<br />
Object tracking which is designed in this final project will identify an object based on color. Identification process of the color element is done by extracting information from objects named Hue. Hue is used as an identification element to accelerate the process of object recognition because hue is a single data channel. Hue of an object will be stored in a memory and will be used as a reference for further process. Tracker that is designed will follow an object that had a same hue's value with hue stored in memory. Back projection is used to re-map the pixels that have the same hue in memory. Results of back projection is a binary image that represents the distribution of pixels of the object being tracked. CamShift algorithm is used to determine the center mass of the pixel distribution continuously. Therefore, object tracking will always be focused to the center of the object which is tracked. Object tracking that is designed will work well on objects that have different colors to its environment. |
|---|