Auto-focus And Yarn Color Measurement Methods In Multi-spectral Imaging System

With the development of the color information technology, the multi-spectral imaging system is widely used in various fields of life and production becomes of its advantages. This paper solves a technical problem, automatic focus, in the multi-spectral imaging system. Also we put forward a new measuring method of yarn’s colors using multi-spectral imaging system.Multi-spectral imaging system acquires spectral color images by splitting the light spectrum into multiple bands using filter. Due to the difference of refractive index with wavelength, the focus positions also are different. So we should adjust each channel in order to obtain a sharp image. If using traditional way to focus, the efficiency is low. To solve this problem, this paper put forward a new method of autofocus by using the similarity of focus measure distribution in each channel. First, select a reference channel and use the traditional way to focus and to get its focus measure distribution. Second, use the similarity of focus measure distribution to find the focus position in other channels. Verified by the experiment, the method can focus fast and accurate.Yarn’s color measurement plays a very important role in the field of textile printing and dyeing. To weave the qualified cloth, we should first get the yarn’s color accurately. The traditional method of yarn’s color measurement has its disadvantages: need a lot of yarns. To solve this problem, this paper puts forward a new method for measuring yarn’s colors using multispectral imaging system, because it can measure the spectral reflectance of each pixel. First take a multispectral image of single yarn. Then use the k-means algorithm to segment and extract the yarn’s color, finally using polynomial model to map the measuring color to true color. The experimental results show that the method can accurately measure the yarn’s color, and overcome the disadvantage of traditional method.