| With the rapid development of the information field,the traditional sensing system gradually falls into a bottleneck due to the limitation of Nyquist sampling theorem in meeting the demands of mass data acquisition,processing,real-time storage and efficient transmission.The proposed compressed sensing theory enriches the theory of signal acquisition and also brings a new idea to solve the problems of transmission efficiency and hardware cost for the research and implementation of image processing and transmission system.Compared with traditional image compression methods,the image compression method based on compressed sensing can establish a data transmission system with lower delay,lower resource loss,higher transmission efficiency and better real-time performance.This paper focuses on the application of compressed sensing algorithm for image signals,and carries out the design and implementation of compressed sensing based image compression transmission system on FPGA hardware platform.In this paper,the image compression processing method based on compressed sensing theory is studied in the following three aspects: the selection of sparse representation method,the construction of measurement matrix and the study of image reconstruction algorithm.Firstly,a sym wavelet transform basis is selected to achieve sparse representation of images by analyzing various sparse representation methods and their sparse base performance.Then,the deterministic matrix is researched for hardware realization and good measurement performance.In this paper,an alternate chaotic measurement matrix based on Lorenz chaotic system and Chen chaotic system is designed,and its constrained equidistant properties,randomness and reconstruction performance are analyzed and verified.Finally,the reconstruction algorithm of compressed sensing theory is studied and analyzed,and OMP algorithm is selected to realize the reconstruction of compressed image.In terms of hardware implementation,the image compression transmission system designed in this paper takes Cyclone IV series FPGA as the core,realizing the functions of image acquisition and processing,dimension reduction and compression,coding and recombination and Ethernet transmission.The image acquisition and processing function is realized based on OV5640 module,and the sequence control and subsequent image framing,Gaussian filtering and format conversion are completed by FPGA.The dimension reduction compression part realizes data compression by the idea of dimension reduction projection operation according to the compressed sensing theory.In this paper,according to the pipeline structure and parallel operation ability of FPGA,a pipeline pulsating row and row compression structure is designed,so that the system can compress the image data by the sensor matrix in line with the row and row sequence.The data after dimension reduction and compression is further compressed by using the parallel processing strategy for Delta encoding and Zigzag encoding subsequent encoding,and then realizes the output of code stream conversion through the encoding recombination module.Compressed image data is cached by DDR2,and then the transmission module completes packet encapsulation and data frame transmission according to the command parsing result.Finally,the performance of the image compression transmission system is analyzed and evaluated.Experimental results show that the system can achieve efficient image compression and transmission,and has a high compression ratio.The compressed image is reconstructed in Matlab,and the quality of the decompressed image is measured by PSNR(peak signal-to-noise ratio)and SSIM(structural similarity).According to the analysis of test data,the quality of reconstructed image is good. |
