The Design And Research Of Embedded Image Acquisiton Terminal Based On Blackfin DSP

Nowadays, along with the increasing demand of image quality and the development of the technology of image compression and signal processing, image data after compressing has so largely decreased the gross amount and transmission bandwidth that image data of high definition could be expediently transmitted in normal communication system, so the applications of image surveillance, such as, image surveillance system based on base station without guard, fingerprint identification system, etc, have been highly extended.The thesis firstly analyzes the current situation and development of the image compression coding technology and embedded system technology, compares different schemes of the system, and finally decides to use the embedded system to implement JPEG still picture compression coding with the wireless transmission to realize the design and research of feasibility of the image acquisition terminal.The thesis introduces interrelated knowledge of digital compression technology and basic manner, and emphatically analyzes the basic principle, realization flow, and correlative key technology of JPEG still picture compression coding.On the basis of the analysis of different design schemes, the thesis chooses the Blackfin 533 digital signal processor of ADI Corporation and uses it as the core to design the overall hardware circuit of the system. Also, this thesis gives details of modules of the main processor, the image acquisition, the memory and the wireless transimission, interfaces of interrelated peripheral circuit, and summarily introduces the important requirement when designing the circuit board. Then the thesis gives the design of the software of the kernel transplantation of the uClinux operation system, the driver of the picture sensor, the optimization of JPEG algorithm, the driver of the wireless transmission and so on. Finally the whole system design has some traits of simple configuration, low power consumption, small volume, and strong expansibility, etc.At the end of the thesis, the part of system testing detailedly introduces testing flow and method, also after testing we can get both the amount of system clock before optimizing and the one after optimizing, and we know that the efficiency of each discrete cosine transform is improved by comparing data of system clock, and that the performance of the whole system is also ameliorated, moreover, we can aquire different pictures which were compressed by adopting different quantization matrix after decoding from personal computer. At last, the thesis concludes and summarizes the process of the whole system design.