The Color Recognition Technology Research And Application With Machine Vision Under Virtual Environment

The color quality examination is getting more and more widespread in the industrial production and the agricultural and subsidiary products processing industry's application, receives more and more takes seriously. Carries on the color quality using the machine vision method the examination to be possible both to be able to guarantee the examination the objectivity, the unity, and can eliminate the noise which well the outside each kind of disturbance factor creates, enables the examination to be able even more precise.The paper research's hard core is the machine vision system's software part - image processing part. In constructs on the good machine vision hardware system, application hypothesized instrument software LabVIEW and NI Vision develop the platform, has realized through the programming to image gathering and processing. First through establishes CCD the parameter and the software programming realizes to the ketchup image gain; Then the application confined smoothing procedures, ultra limit the domain law and the value filtering to the ketchup color image which gathers carry on the smooth filter, eliminates the external interference and the noise; In the next application threshold value division law's most superior threshold value division method and the Sobel marginal check operator carries on the ketchup image the division, will use in the image the ketchup color RGB tertiary color value dividing; Uses the bending moment theory not to withdraw the ketchup color RGB tertiary color value, finally unifies the BP neural network algorithm to give the ketchup color quality test result.