Color Gradient Model for Object Detection and Tracking System
In recent years object detection and recognition methods getting popular and often requires high accuracy rate
since it happening in outside environment and also because of non linear trajectories. In most cases local features based on
object shape or textures etc .are used. Among them COLOR DERIVATIVE feature is typical. Many papers have been
published in Color Self-Similarity (CSS) driven color information. CSS involves computing color similarity and gradient
model that represent a feature that is effective for object detection. It has also been reported that detection performance can
be improved by combining the CSS feature together with the COLOR DERIVATIVE feature or other shape-based feature.
We propose a Color gradient feature that is based on color similarity for object detection and tracking system.