An Investigation of Image Segmentation of Building Using Wavelet Transform

• Main Authors: Cho-yang Chen, 陳卓揚
• Other Authors: Ko-Young Shao
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/21014970369743885233

碩士 === 義守大學 === 土木與生態工程學系碩士班 === 93 === This paper proposes the limitation and the usage of the wavelet transformation on one-dimensional and two-dimensional signal analyses. An investigation of the wavelet transform on noise image, building image, and different mother wavelets was presented. Using image segmentation on automatic construction could lead to the optimization of the automation system of construction. This one-dimensional signal, noised signal, probing into the wavelet transform, and transforming wavelet correctly is analyzed in this research. The wavelet transform and the inverse wavelet transform exhibit little variations on the signal. The wavelet transform analyses on noise images are more efficient than traditional methods. The using of wavelet transform on two-dimensional noise image can be achieved by extending the wavelet transform on one-dimensional noise image. This paper proposes that the wavelet transform can be used to filter out noise. However, using the wavelet transform on image segmentation could derive better results.