基于粒子群算法的番茄酱成分检测及品质分析

Component Detection and Quality Analysis of Ketchup Based on Particle Swarm Optimization Algorithm

以番茄酱成分检测过程为研究对象,采用多光谱成像技术采集番茄酱光谱图像,并提取出有效光谱信息,分别利用偏最小二乘法、偏最小二乘法支持向量机算法和粒子群算法建立番茄酱成分校正模型和番茄酱品质预测模型。采用3种模型进行预测试验对比,偏最小二乘法支持向量机算法和偏最小二乘法两种模型对可溶性固形物和番茄红素含量的预测性能均低于粒子群算法模型;粒子群算法模型对可溶性固形物和番茄红素含量的预测绝对系数均大于0.9,表明粒子群算法模型能够有效进行番茄酱成分含量检测和品质预测。

The detection process of ketchup components is taken as the research object,the multispectral imaging technology is used to collect the spectral images of ketchup,and the effective spectral information is extracted.Partial least squares method,partial least squares support vector machine algorithm and particle swarm optimization algorithm are used to establish ketchup composition correction model and ketchup quality prediction model respectively.The results show that the partial least squares support vector machine and partial least squares are better than particle swarm optimization algorithm in predicting the content of soluble solids and lycopene.The predicted absolute coefficients of the content of soluble solids and lycopene by particle swarm optimization algorithm model are all greater than 0.9,which indicate that the particle swarm optimization algorithm model could effectively detect the content of ketchup components and predict its quality.