Multi-channel Image Acquisition Preprocessing System Based On Dsp

Nowadays, with the rapid development of imaging technology, many imaging technology hardware electronics products continue to emerge. At the same time, daily consumption and scientific research of the image acquisition system requires for high speed and expandability, single-chip as a core processor can not satisfy the real-time performance and scalability requirements. Therefore selecting a suitable image processor for image acquisition system is very important. Specifically for high-speed digital processing and the design of digital signal processor, DSP has become a figure role of image processing area, at the same time the expansion of the capacity of DSP is able to satisfy the various needs of modern field as image acquisition system. In this paper, a multi-channel image acquisition system with the core processor of DSP was designed.The system is designed for the feature point extraction of real-time and three-dimensional objects. It provides the next match of stereo vision and three-dimensional reconstruction for image process-related information. Therefore multi-channel image acquisition system does not only have to ensure the real-time image acquisition, acquisition quality and stability, but also the function of network transmission, transfers the processed image information to PC.Based on the above needs, the multi-channel image processing system based on DSP is designed as below:1.Summed up the technology image acquisition system, and introduced network function used in image acquisition system.2.Discussed the hardware design of the multi-channel image acquisition system. Designed the image acquisition system in a modular way separately, and including image acquisition module, network module, memory module etc.3.Designed a real-time operating system kernel based on embedded system software program. The program used TI provided RF-5 software reference framework, including the preparation of hardware drivers, the implementation of the mission modules and systems integration, then system optimization was performed about software and hardware.4.For the system, Harris feature point extraction algorithm was used and its performance and the principles were introduced. Then put forward the advanced algorithm according to application.5.By using the application Development System NDK network components of network communication and network applications for mission design, the network transmission was completed.