COST-EFFECTIVE AUTOMATIC ROI DETECTION FOR REAL TIME IMAGE PROCESSING

Lucas Leiva

Abstract


The automatic region of interest (ROI) detection of an image is a vital component in visual pattern recognition systems. Through this technique, it is possible to analyze only those regions that contain relevant data. Several methods were described in the last years for this purpose. Corner detection is one of several algorithms that can be applied to solve the problem, detecting the salient points of an image. The regions that contain salient points will be marked as ROIs. In this work a technique based on an optimization of FAST corner detection is used for automatic region extraction. The proposed algorithm (Express FAST) is intended to be implemented in hardware architecture and thus, both time and area are critical factors. The provided solution is also compared to the results of another corner (and subsequently ROI) detector, and the results showed satisfactory results. 


Keywords


image analysis, region of interest detection, salient points, video processing.

Full Text:

PDF

Refbacks

  • There are currently no refbacks.