The Research And Design Of The Color Difference Detection System Based On DSP For Dyeing Products

The printing and dyeing industry is the main industry of Zhejiang Province,and the product quality is directly related to the position of the industry in the international competition.The color is the most important quality index of printing and dyeing products,and the color quality control is closely related to the color difference testing equipment.Currently,the devices of the color difference detection are mainly based on PC,by using the powerful data processing capabilities of PC to realize the color difference detection algorithm,but its occupied space is relatively large,so is its cost.As for image processing,the professional DSP has a powerful data processing capabilities,and its volume is relatively smaller,and its cost is less.The key technologies in the field of color difference detection are the evaluation algorithm of color difference and the illumination correction algorithm,the illumination correction algorithm is mainly to solve the problem of the evaluation of the color difference under different light conditions and uneven illumination under the same light source,while the evaluation algorithm of the color difference is mainly in order to establish a model of the color characteristic information and the evaluation indexes,so as to make reasonable judgment in accordance with the characteristics of the color information and the output of the corresponding action.The point of view of this thesis is based on machine vision,using the DSP chip which can makes the system of the color difference detection miniaturization and facilitation.At first,after building the image acquisition system based on DSP,and then using the DSP for image pre-processing,illumination correction and color difference evaluation for the collected images,the corresponding algorithms are developed,simulated and optimized by using Matlab on PC,and finally transplanted to DSP system.For image pre-processing,the adaptive gaussian filter is put forward in this thesis.By experimental contrast,it can be seen that the filter have better filtering effect.For illumination correction algorithm,the uniformity illumination correction algorithm under the same light source and the illumination correction algorithm between different light sources are proposed in this thesis.The uniformity illumination correctionalgorithm is based on brightness,where it is can be seen that the effect of the uniformity illumination correction algorithm is more in line with visual effect.It is mainly based on the Minimum Support Vector Machine algorithm and the combination of gray theory algorithm between different light sources.By compared with SSR and BP_Adaboost,we can conclude whether the correction effect is good or not.For the evaluation of difference color,this thesis mainly uses the Minimum Support Vector Machine algorithm in two aspects of the color uniformity and consistency.Regarding CIEDE2000 color difference formula as reference,this thesis improves the optimization of CIELAB color difference formula to gain accurate and simplified color difference formula. Compared with the standard visual grade,it shows that the color difference evaluation algorithm has better evaluation result.