Multi-connected visual processor and image processing method using the same

Abstract

Disclosed are a multi-connected visual processor and a processing method thereof. The visual processor comprises a reduced instruction set computing (RISC) processor module, a system bus module, an image data storage device module, a picture element (PE) processor array module and a row processor (RP) array module, wherein all of the modules are connected with the system bus module, and the reduced instruction set computing processor module is capable of controlling the operation of other modules through the system bus module; the image data storage device module, the picture element processorarray module and the row processor array module are sequentially connected; particularly, muti-connected design is adopted by the picture element processor array module and the row processor array module, image statistical characteristics calculation acceleration is facilitated, and statistical speed of image histograms is improved to the most extent. The processor has the advantage of being simple in structure, and the processor can be applied in the occasion where high-speed image processing is needed.

Description

technical field [0001] The invention relates to the technical field of visual image processing such as a programmable visual chip and a parallel visual image processor, in particular to a multi-interconnected visual processor and an image processing method using the same. Background technique [0002] The traditional visual image processing system includes a separate data source and a general-purpose processor. The data source is used to obtain the original image data, and the general-purpose processor performs corresponding processing. The general-purpose processor is generally a CPU, and image processing is often performed serially pixel by pixel. In practical applications, when there are many image processing operations, it cannot meet the requirements of real-time processing; in order to improve the processing speed, GPU is often used instead The CPU performs image processing, but the visual image processing system with GPU is not portable enough to be used in occasions ...

AI Technical Summary

Problems solved by technology

However, this multi-level parallel visual processor simply accelerates calculations by stacking hardware. Many image processing involves calculating statistical features. When this multi-level parallel visual processor calculates the statistical histogram, although two-dimensional Hardware accelerated, but statistical histograms still take most of the image processing time

Images