Research On Denoising Packaging Image By Wavelet Transform

A packaging image is often disturbed by imaging equipment and outer environment noise while it is sampled, transformed and transmitted. Thus the packaging image quality declines. If you don't know the accurate position and the scope of noise in a packaging image, you won't recover it into original image. Generally speaking, noise appears in a packaging image randomly.Although noise appears in a packaging image randomly, it has a number of disciplinarians. Therefore it is feasible to remove noise at a certain extent. Because noise and original packing image amalgamate, it is not avoidable to destroy some details of the original packing image by using any method to remove noise drastically while you don't add some quantity of information into the original packing image. So for a packing image with noise, you can't find arithmetic which can not only remove noise drastically but also reserve all details of the original packing image.To restrain noise and improve the packaging image quality and convenient for continuing to process, packing images must be denoised. Because wavelet has many advantages such as multi-resolution, low-entropy, agility to choose radicle and removing relativity, you can gain satisfied effect by using wavelet method to denoise for a packing image.The aim of this subject is to develop software for removing noise from a packing image by using wavelet method after wavelet method is studied carefully. Because a series of wavelets are used in this software, you can use this software to denoise for a packing image neatly. And you will gain different effects if you choose different wavelets, different wavelet-decomposed progressions or different wavelet-reconstructed thresholds.