Research On Image Compression And Signal Detection Based On Wavelet Transformation

With the great development of information technology, it is the trend for the industry process control and the secure surveillance of some important departments to realize image surveillance. However, the image data is giant, the transmission of image data is the key section of image surveillance system. This paper, based on the project of the image supervisor system of the substation of QuZhou electrical power company, researches the technologies of image compression and coding, and studies the methods of signal detection from image data as well. The primary achievements are following:1. Regularity of wavelet bases and wavelet filters on image wavelet transform for compression and coding is studied. The computerization experiment results showed that, for a flat image, the higher the regularity order of wavelet bases is, the more concentrated is the information, therefore it can get greater compression ratio; but, for a cross-grained image, this is not the case.2. Boundary processing and the phase shift of wavelet filters are also studied. The author considers that forward-period extension in decomposition and backward-period extension in reconstruction of signal is a good strategy, it can keep the length and phase shift of the signal unchangeable.3.The techniques of image compression and coding based on wavelet transform are discussed; they mostly involve wavelet coefficients EZW (embedded zero-tree wavelet) coding and wavelet coefficients vector quantization coding. For main work, the application of image wavelet transform coding to engineering is studied; some technical details are analyzed and simulated.4. Fractal image compression is one based on the theory of fractal geometry, which has high image compression ratio, however, its algorithm is complex and running efficiency is low. In this paper, a method of fractal image compression and coding based on Genetic Algorithm (GA) is proposed. It can much reduce theprogram running time and that can keep the same quality of restoration image and compression ratio as the general method. On the principle of GA, this paper improves the structure principle of GA searching space; it can avoid the production of super children and the loss of some excellent parents in the GA searching. Experiments confirm the efficiency of the method.5. Signal detection is the key part of control system, but sometime, the universal method based on the transducer is disabled in the some cases. The signal detection based on image is increasingly taken to account in engineering with the development of image processing and computer technology. Since the wavelet analysis is a local time-frequency analysis, and is praised "mathematics microscope", it has good effect on the detection of signal singularity. This paper investigates the techniques of signal detection based on wavelet transform and discusses the inner relationship between the character of signal singularity and modular maximum of wavelet transform, and presents a new method of detection welding slam based on the wavelet transform as well. Experimental results demonstrate its feasibility.