GPU-accelerated image alignment for object detection in industrial applications

Abstract

This research proposes a practical method for detecting featureless objects by using image alignment approach with a robust similarity measure in industrial applications. This similarity measure is robust against occlusion, illumination changes and background clutter. The performance of the proposed GPU (Graphics Processing Unit) accelerated algorithm is deemed successful in experiments of comparison between both CPU and GPU implementations

HS‐IDEOR—A tool for conceptual modeling and designing of hardware architectures focused on object recognition

“…Research by [7] and [10] suggested that a graphics processing unit (GPU) would be a better option when training is required to implement algorithms for object recognition due to its optimization for parallelization of mathematical functions using thousands of small cores. Unfortunately, the application area of GPUs is limited because of the need for a processor-based system.…”

“…Furthermore, the training of undergraduate students in this area typically follows a sequential approach to implementing object recognition algorithms as a sequence of instructions using a personal computer, where the central processing element is a processor. Research by [7] and [10] suggested that a graphics processing unit (GPU) would be a better option when training is required to implement algorithms for object recognition due to its optimization for parallelization of mathematical functions using thousands of small cores. Unfortunately, the application area of GPUs is limited because of the need for a processor‐based system.…”