The structure of the image processing and recognition module is related to the overall performance of the system. The FPGA+DSP architecture is conducive to building an efficient data processing process and facilitating the allocation of processing tasks, and improving the parallelism and resource utilization of the system. SRAM, SDRAM and FLASH in the system are directly connected to DSP for use: FLASH is used to store programs and some fixed table data; SDRAM, as the system memory of DSP, is used to run system programs; SRAM is a high-speed data storage area used to store temporary variables generated by program operation. DDR SDRAM is specially used to store some high-capacity data blocks, such as fingerprint data collected and pixel gradient data calculated during preprocessing. It is directly connected to FPGA and is the fastest memory area in the system. In addition to serving as the expansion bus interface of DSP processor, FPGA also shares some data processing tasks, because only one DSP is not competent for all computing and control tasks. When processing fingerprint data, it often encounters some tedious addition and subtraction operations and logic operations. Usually, this part is processed by FPGA. Considering the particularity of fingerprint processing algorithm, it is also necessary to realize DDR control function.

    Due to the large amount of mathematical operations in the process of fingerprint identification, the program design inevitably requires a large storage space. In order to improve the overall performance, the heavy computing tasks need to be handed over to the DSP for processing, while the image acquisition part needs to occupy as little DSP time as possible. In addition, using the interval of image acquisition or at the same time of image acquisition, hardware can complete some simple and tedious operations to share the processing tasks of DSP, improve the parallelism of processing, and meet the requirements for real-time. The system uses TMS320VC5402, which has a fast computing speed and high cost performance. The 8bits gray scale fingerprint image collected in the system occupies one byte per pixel, and the image size is 512 × 512 pixels in size, 256k bytes of storage space is required to store a frame image. DSP unit is the core of the whole fingerprint processing system, which is responsible for real-time fingerprint processing.