Digital image edge plays an important role in image analysis, such as image segmentation. The edge includes the valuable information of the image which can be used in image understanding and analysis. And through edge detection, we can greatly reduce the calculation of image analysis and processing in the following step. With the method based on wavelet, we developed a method for times series digital images. Our method exploits the fact that apparent edges are present on the brightness time series curve of some pixels corresponding to sharp and abrupt transitions of brightness. Through a wavelet transform of time series curve of pixel, our method identifies and tracks significant edges on the curves. Through variance analysis and The Bayes Classifier, an optimal edge strength threshold is statistically determined to differentiate real edges from weak edges caused by noisy perturbations. At the end of the paper, we improve the method of searching the edge strength threshold, which increase efficiency of digital image classification. |