Improved method for image thresholding based on fuzzy index
To solve the problem of automatically getting the window size with traditional segmentation method, an adaptive method for image thresholding based on index of fuzziness is proposed by combing pre-segmentation with histogram information. Furthermore, in order to segment images with one mode histograms by image thresholding based on index of fuzziness, a new smooth iteration formula for image enhancement with neighborhood information of pixels is proposed. Through the iteration formula and adaptive method for image thresholding based on index of fuzziness, the applicable range of image thresholding based on index of fuzziness is expanded. Experiment result shows that with this method, the accuracy of target segmentation can be up to 97.3% and the performance shows good robustness.
NETL
NSTL
万方数据
维普
