基于支持向量机模型和图像处理技术的甜椒叶面积测定

叶片是作物进行光合作用的重要器官，是研究作物对光能吸收的一个主要的生物学指标。应用支持向量机理论，建立了支持向量机叶面积模型，输入参数为叶片长度、叶片最大宽度，输出参数为叶面积。通过对应用计算机图像处理技术测量所得到的样本数据进行训练，以叶片的长度、宽度作为输入参数对叶面积进行模拟及检验，并将模拟结果与线性回归和人工神经网络模型进行了对比。结果表明，支持向量机叶面积模型的最大误差为6.09%，平均误差为2.73%，模拟精度达到0.996。该方法能较真实地反映甜椒叶面积的实际大小，具有较好的实用价值和应用前景。

Determination of leaf area of sweet pepper based on support vector machine model and image processing

As a vital organ for crop photosynthesis, leaf is one of the major biological indicators in the study on light absorption by crops. Support Vector Machine (SVM) theory was used to set up a SVM model for determination of leaf area of sweet pepper, the input parameters were the leaf length, maximum width of the leaf, and the output parameters were the leaf areas. Data measured by computer image processing technology were trained as samples, the length, maximum width of the leaf were used as input parameters to simulate and test the leaf area. The results were compared with those of linear regression and artificial neural network model. The results showed that the maximum error of leaf area determined by support vector machine model was 6.09%, and the average error was 2.73%, the simulation accuracy was 0.996. This method can well reflect the actual size of leaf area of sweet pepper, and has good practical value and application prospect.