Acceleration corner detection method and mobile device for implementing acceleration corner detection method

Abstract

The invention provides an acceleration corner detection method and a mobile device for implementing the acceleration corner detection method. The acceleration corner detection method is implemented through a graphic chip on the mobile device and comprises the following steps: the graphic chip reads a two-dimensional image and conducts corner detection on the two-dimensional image in a multi-core parallel computing mode according to a second program code which is written with an open graph program library shading language so as to find out corners in the two-dimensional image. According to the acceleration corner detection method and the mobile device for implementing the acceleration corner detection method, the characteristic of multi-core parallel computing through the graphic chip and a multi-core corner detection execution algorithm are utilized for simultaneously detecting multiple image points so as to conduct corner detection, and the speed of image identification can be increased for the mobile device.

Description

technical field [0001] The invention relates to a method for accelerating corner point detection, in particular to a method for accelerating corner point detection by using a graphics chip on a mobile device and a mobile device for executing the method. Background technique [0002] In image recognition, corner detection is a very important technology, and is widely used in various image processing related applications. Corner detection is used in applications such as 3D reconstruction, object recognition, image mosaicing, and motion tracking. With the popularization of mobile devices, image recognition technology is often applied on mobile device platforms. [0003] Since the launch of the iPhone in 2008, smart mobile devices have become an indispensable product for people. Since mobile devices are close to people's lives and have the characteristics of being portable, the application of image recognition on mobile devices is also more diversified, such as using face reco...

AI Technical Summary

Problems solved by technology

[0006] However, these methods of using GPGPU acceleration are only applicable to the latest (future) three-dimensional (3D) graphics chips, such as the NVIDIA GeForce GTX series, and are currently only available on the p

Images