小波变换在果品图像处理中的应用

Wavelet Transform and Its Application in the Processing of Fruit Image

针对果品图像处理常用方法不能同时在时域和频域分析图像,且不具有多分辨率特性的问题,应用小波变换理论和技术,以红枣图像为例,对果品图像进行了去噪、增强等处理。小波去噪所需时间为9s,还不及数学形态学的1/4;小波变换用于红枣图像增强所用的时间仅为7s,是模糊数学形态学的1/495。试验结果表明:小波变换用于果品图像增强和消噪,具有方便快捷、去噪效果好、目标明确等优点;小波变换用于果品图像处理是有效的、可行的。

As a traditional method of image processing can't analyze image in the domains of time and frequency simultaneously and it hasn't the characteristic of multiresolution, in this paper taking Chinese date image as a instance, we used wavelet transform to process, denoise, and enhance the fruit image. Compared to the mathematical morphology, wavelet transform spent 9 seconds to denoise the image, which was less than a quarter of a mathematical morphology demand. Compared to the fuzzy mathematical morphology, wavelet transform spent only 7 seconds to enhance the image, which was 1/495 of the fuzzy mathematical morphology demand. The experimental results show that the wavelet transform technique used in fruit image enhancement and de-noise is effective and practicable.