目的利用近红外光谱及光谱融合策略,结合化学计量学方法,建立水性油墨的颜色预测模型,实现水性油墨印刷品颜色准确预测。方法采集不同酒精含量和不同调色墨含量的油墨的近红外光谱反射率和吸光度数据,并测得对应的印刷品的Lab值,然后建...目的利用近红外光谱及光谱融合策略,结合化学计量学方法,建立水性油墨的颜色预测模型,实现水性油墨印刷品颜色准确预测。方法采集不同酒精含量和不同调色墨含量的油墨的近红外光谱反射率和吸光度数据,并测得对应的印刷品的Lab值,然后建立单一光谱不同预处理过后的偏最小二乘(Partial Least Squares,PLS)模型,以及基于数据层融合和特征层融合的PLS模型,最终通过比较预测集决定系数和预测集均方根误差(Root Mean Square Error of Prediction,RMSEP)以及色差来评估模型的预测效果。结果单光谱建模,基于反射率建立的模型准确率高于基于吸光度建立的模型;数据层融合缺乏稳定性,对L和b值的预测有所提升,对a值的预测几乎不变;特征层融合建模效果明显好于单一光谱和数据层融合,对Lab的预测决定系数分别达到了0.9961、0.9939、0.9974;RMSEP值分别为0.1421、0.2126、0.2072;预测值与真实值的最大色差为0.6783。结论通过光谱特征融合技术能提高油墨颜色预测精度,准确预测出油墨颜色变化。展开更多
Based on plastic bending engineering theory and machine vision technology, the intelligent control technology for forming steel pipe with JCO process is presented in this paper. By ‘twice pre-bending method’ in the ...Based on plastic bending engineering theory and machine vision technology, the intelligent control technology for forming steel pipe with JCO process is presented in this paper. By ‘twice pre-bending method’ in the first forming step, the springback law can be obtained. With the springback law and the target angle, the exact punch displacement which determines the formed angle in each bending step is predicted. In the succedent forming steps, the bending process is carried out with the exact punch displacement by real-time revising the springback law. And the angle error in each forming step is calculated by comparing the actual formed angle with the target angle. By conducting compensation for the last angle error in the next forming step, each precise bending process step is realized. A system of intelligent control technology for forming the steel pipe was developed. A calibration method is proposed to calculate the exterior parameters of the CCD camera, in which the equilateral triangle is em-ployed as the calibrating board and only one image needs to be captured. A mathematical model, which converts the angle in the image into the actual formed angle, is derived. The experimental results showed that the ellipticity of the formed pipes was less than 1.5% and the high-quality pipes can be manufactured without the worker's operating experience by employing the in-telligent control technology.展开更多
文摘目的利用近红外光谱及光谱融合策略,结合化学计量学方法,建立水性油墨的颜色预测模型,实现水性油墨印刷品颜色准确预测。方法采集不同酒精含量和不同调色墨含量的油墨的近红外光谱反射率和吸光度数据,并测得对应的印刷品的Lab值,然后建立单一光谱不同预处理过后的偏最小二乘(Partial Least Squares,PLS)模型,以及基于数据层融合和特征层融合的PLS模型,最终通过比较预测集决定系数和预测集均方根误差(Root Mean Square Error of Prediction,RMSEP)以及色差来评估模型的预测效果。结果单光谱建模,基于反射率建立的模型准确率高于基于吸光度建立的模型;数据层融合缺乏稳定性,对L和b值的预测有所提升,对a值的预测几乎不变;特征层融合建模效果明显好于单一光谱和数据层融合,对Lab的预测决定系数分别达到了0.9961、0.9939、0.9974;RMSEP值分别为0.1421、0.2126、0.2072;预测值与真实值的最大色差为0.6783。结论通过光谱特征融合技术能提高油墨颜色预测精度,准确预测出油墨颜色变化。
基金Supported by the National Natural Science Foundation of China (Grant No. 50805126)the Hebei Natural Science Foundation (Grant No. E2009000389)
文摘Based on plastic bending engineering theory and machine vision technology, the intelligent control technology for forming steel pipe with JCO process is presented in this paper. By ‘twice pre-bending method’ in the first forming step, the springback law can be obtained. With the springback law and the target angle, the exact punch displacement which determines the formed angle in each bending step is predicted. In the succedent forming steps, the bending process is carried out with the exact punch displacement by real-time revising the springback law. And the angle error in each forming step is calculated by comparing the actual formed angle with the target angle. By conducting compensation for the last angle error in the next forming step, each precise bending process step is realized. A system of intelligent control technology for forming the steel pipe was developed. A calibration method is proposed to calculate the exterior parameters of the CCD camera, in which the equilateral triangle is em-ployed as the calibrating board and only one image needs to be captured. A mathematical model, which converts the angle in the image into the actual formed angle, is derived. The experimental results showed that the ellipticity of the formed pipes was less than 1.5% and the high-quality pipes can be manufactured without the worker's operating experience by employing the in-telligent control technology.
