OpenGL ES (open graphics library for embedded system) implementation method for ray casting algorithm under ARM+GPU (advanced RISC machine+graphic processing unit) heterogeneous architecture

Abstract

The invention discloses an OpenGL ES (open graphics library for embedded system) implementation method for a ray casting algorithm under an ARM+GPU (advanced RISC machine+graphic processing unit) heterogeneous architecture. The method comprises the following steps of 1, establishing a display window system in an ARM; 2, acquiring initial body data from the ARM, and recreating the initial body data to a three-dimensional Cartesian coordinate system; 3, arranging two-dimensional textures by using the ARM, and importing three-dimensional ultrasonic body data in a two-dimensional texture way by using an OpenGL ES interface; 4, receiving a human-computer interaction instruction by using a GPU, and calculating a coordinate transformation matrix to realize a human-computer interaction function in three-dimensional imaging; 5, performing OpenGL ES implementation on the ray casting algorithm by using the GPU, and controlling the calculated amount; and 6, drawing a display plane by using the GPU. According to the method, a conventional ray casting algorithm based on a graphics card and an Intel processor is implemented under an embedded platform, so that the system development cost is lowered.

Description

technical field [0001] The invention relates to the realization process of three-dimensional imaging in the field of image processing and computer vision in an embedded system, in particular to the OpenGL ES realization of a ray projection algorithm under the ARM+GPU heterogeneous System On Chip (SOC) architecture. Background technique [0002] Medical ultrasound imaging is a medical imaging diagnosis technology based on ultrasound. It sends ultrasound to the human body through an ultrasound transducer and receives tissue echoes. It uses the physical characteristics of ultrasound and the differences in the acoustic properties of human tissues and organs to display the morphology of the human body. information for disease diagnosis. It has the advantages of no trauma, no radiation, low cost, fast imaging speed, etc., and it is portable and can acquire images in real time. Therefore, ultrasound imaging occupies a large proportion in general medical diagnosis. In particular, f...

AI Technical Summary

Problems solved by technology

[0008] However, for the heterogeneous architecture of ARM+GPU, embedded GPU has two obvious limitations: 1) less parallel computing units and weak computing power; 2) ARM and GPU share system memory, and there is no dedicated high-speed memory

At the same time, under the embedded system, the dedicated open graphics library interface has also changed from OpenGL to OpenGL ES. Therefore, the mature ray-casting algorithm developed under OpenGL cannot be directly applied to the embedded system.

Images