Multi-Resolutional Image Analysis by using Wavelet Transform

使用小波变换进行多分辨率图像分析

• 批准号: 08650487
• 负责人: EIHO Shigeru
• 金额: $ 1.47万
• 依托单位: KYOTO UNIVERSITY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 1996
• 资助国家: 日本
• 起止时间: 1996 至 1997
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-08650487/
• 关键词: wavelet transform; detection of defect; frequency analysis; 布の織り傷検査; 画像処理; 非線形画像フィルタ

We need appropriate pre-processing for measuring or detecting various kinds of information from image by the digital image processing such as the noise reduction and the edge enhancement. Some powerful methods according to the above processing are the methods in the frequency domain such as the Fourier transformation. They need more computation time and memories than the spatial methods which are also easy to catch the meaning of the processing. The aim of this research is to develop the system of multi-resolutional image analysis by using the wavelet transform which will need less time for computation and have clear information of the signal on the position in the image space. We applied the developed system to detect the defect of the product sheets of the cloth (fabric) and metal. As the results, the defect parts are detectable by the multi-resolutional method developed by using the wavelet transform. As the frequency components of the wavelet transform are calculable with the neighbor points of the image positions, the method developed is applicable as the real-time processing of the data and on-line detection of the defects.

在数字图像处理中，为了测量或检测图像中的各种信息，需要进行适当的预处理，如降噪、边缘增强等。根据上述处理的一些强大的方法是频域中的方法，例如傅立叶变换。它们比空间方法需要更多的计算时间和内存，空间方法也容易捕捉处理的意义。本研究的目的是利用小波变换建立一个多分辨率图像分析系统，该系统计算时间短，信号在图像空间的位置信息清晰。我们应用所开发的系统来检测产品的布（织物）和金属片的缺陷。结果表明，利用小波变换的多分辨率方法可以检测出缺陷部位。由于小波变换的频率分量可由图像位置的相邻点计算，因此该方法适用于数据的实时处理和缺陷的在线检测。