【作者】 段金辉；

【导师】 屈星；

【作者基本信息】 甘肃农业大学 ， 蔬菜学， 2010， 硕士

【摘要】 随着农业现代化进程不断加快,数字农业作为农业生产现代化最前沿的发展领域,有效促进农业生产和现代化信息技术的有机结合,给农业生产发展注入新鲜血液。本研究根据我国数字农业建设的现状以及农业建设与管理的需要,结合甘肃省河西地区加工番茄产业的具体需要,综合运用计算机视觉技术、数学模型分析及肥料效应等方法,为加工番茄番茄红素含量快速检测、积累特性模拟以及氮磷钾肥效应的影响提供理论依据。主要结果如下：(1)利用计算机视觉技术快速测定加工番茄番茄红素含量的方法,建立根据加工番茄果实颜色特征确定其番茄红素含量的回归模型；结果表明：RGB颜色系统的R、G、B、R/G、R/(G+B)、色度坐标r、g、b及HIS颜色系统H、I值与番茄红素呈极显著非线性相关,可用于测定加工番茄果实中番茄红素含量。从建立的4组模型中筛选出拟合度较高的1组模型进行检验,拟合度R2为0.941。使用预测精度较高的RGB颜色特征预测番茄红素含量的模型为:LC=9.0407-0.08G+1.211X-9.882Y+0.0002G2+0.077X2+6.564Y2。(2)为准确了解加工番茄番茄红素积累特性,以10个加工番茄品种为材料,研究筛选适合描述加工番茄番茄红素积累特性的生长模型,用Richards、Logistic和Gompertz方程拟合加工番茄番茄红素积累特征曲线。研究结果表明,Richards方程能够较好地拟合不同加工番茄品种的番茄红素积累特性,其拟合度优于Logistic和Gompertz方程。不同加工番茄品种间番茄红素积累特性的比较研究表明,果实番茄红素含量的变化过程基本可以划分为初始积累、快速积累和稳定积累3个阶段,番茄红素积累主要集中在后2个阶段。提高番茄红素含量可以通过提高番茄红素积累速率和延长番茄红素积累的持续时间这2个策略。(3)采用3414配方施肥设计方法,建立氮、磷、钾肥对加工番茄番茄红素含量的综合作用模型,探讨氮、磷、钾肥对加工番茄番茄红素含量影响的规律。氮、磷、钾肥对加工番茄番茄红素含量的单因素效应、双因素互作效应受到其它因素水平的影响。氮肥对番茄红素含量的影响为正效应,其增加幅度随氮肥施用水平的增加而提高。磷肥对番茄红素含量的影响均为负效应。钾肥对番茄红素含量表现为正效应。当氮肥施肥量为3.27-28.77kg/亩,磷肥施肥量为7.74-22.7kg/亩,钾肥施肥量为3.63-7.57kg/亩时,加工番茄果实番茄红素含量有95%的可能高于14mg/100g。更多还原

【Abstract】 Digital agriculture has been the frontier realm of the modernization of agricultural production since the speed up of modern agriculture development. And it facilitates effectively the integration of agricultural production and modernized information technique, and adds fresh blood to the agricultural production. Based on the current state of the construction of digital agriculture and the need of construction and management of agriculture in China, combining with the practical demands of the processing tomato industry of Hexi Corridor of Gansu Province, Comprehensive use of computer vision technology, mathematical model analysis and fertilizer effects and so on, for the processing of tomato fruit lycopene content in rapid detection, and accumulation of simulation and effect of NPK fertilizer to provide a theoretical basis.The main results are as follows:(1) The rapid methods detecting lycopene concentration by the computer vision technology, and a unary quadratic model to predict lycopene content based on color parameters of processing tomato fruit images have been established in this study. The images of processing tomato fruits were taken in the image cameras, then the color characteristics were extracted with the PhotoshopCS4 image processing software. The correlation between color parameters of processing tomato digital image and lycopene content of processing tomato fruit were analyzed by regression models. The results showed that the color characteristics such as R、G、B、R/G、R/(G+B)、r、g、b in the RGB color system, and H、I in the HIS color system were significantly correlation with lycopene content of processing tomato fruit at P<0.01. Four sets of prediction model were established and among them 1 model with high fitting degree were selected to use. The prediction accuracy of the selected model were tested, and good of fit value 0.941. According to the predict lycopene content of processing tomato fruit, the corresponding model is:LC= 9.0407-0.08G+1.211X-9.882Y+0.0002G2+0.077X2+6.564Y2。(2) To investigate accumulation characteristics of lycopene in different processing tomato varieties and screen a suitable growth equation to describe the lycopene accumulation, ten varieties were used and tested by the Richards, Logistic and Gompertz equations. The results showed that the Richards equation was more suitable for simulating lycopene accumulation process than the other growth equations in processing tomato. The Richards could improve fitting effect among the three models. Via comparing the accumulation characteristics of lycopene content in different processing tomato cultivars, the accumulation of lycopene in processing tomato could be divided into three stages, which were initial accumulation stage, fast accumulation stage and steady accumulation stage. There were two ways for processing tomato cultivars to increase lycopene content:increasing the rate of lycopene accumulation and extending the time of lycopene accumulation process.(3) The action model was established for the effects of nitrogenous, phosphatic and potassic fertilizer on the content of lycopene by using "3414" fertilizer experiment design. The effects of nitrogenous, phosphatic and potassic fertilizer on the content of lycopene in processing tomato were discussed by this model. The single factor and two-factor interaction effects of nitrogenous, phosphatic and potassic fertilizer on the lycopene content were affected by other factors. Nitrogen fertilizer has produced a positive effect to lycopene content, increasing rate became quickly as nitrogen fertilizer increased. whereas phosphor fertilizer has produced a negative effect to lycopene content, Potassic fertilizer has produced negative effect to lycopene content. Fertilizing measure to getting lycopene contents over 14 mg/100g with 95% possibility comprised nitrogen fertilizer 3.27-28.77(kg/667m2), phosphor fertilizer 7.74-22.7(kg/667m2)and potassic fertilizer 3.63-7.57(kg/667m2).更多还原