基于图像处理的油菜分枝及角果着生角度测量方法

摘要

以数字图像处理技术为基础，建立了一套油菜株型角度测量方法，并根据分枝角度将22份材料分为紧凑型（小于30°）3份，中间型（30°~40°）8份和松散型（大于40°）11份；根据角果着生状态将22份材料分为平生型（趋近于90°）3份和斜生型（小于80°）19份。对其中5份材料进行数字图像处理测量值（x）和传统的手工方法测量值（y）比较，经显著性t测验和方差分析，测定结果符合曲线为y=1.0191x-1.3632，相关系数为0.9919。可见，这一套数字图像技术具有较高的重复性、精确性，更具有较好的可操作性，可为油菜分枝角度和角果着生角度的遗传研究提供技术支撑。%Branch and pod angles are important plant architecture traits,which are closely related to ideal plant type,yield,high-density cultivation and mechanical harvesting. For simple and accurate determination of these angles,digital image method was used to angles measurement for rapeseed. According to branch angles,22 materials were measured and grouped into 3 typical types:compact( ﹤30 ° ,3 ),loose( ﹥40 ° ,11 )and interme-diate(30°-40°,8). According to pod angles,the materials were grouped into 2 types:horizontal(close to 90°, 3)and inclined( ﹤80°,19). To test the accuracy of this method,5 materials were studied by comparing with the traditional manual measurements. T test and ANOVA analysis showed a high conformity on angle distinguishing. The high consistency of standard curve was y=1 . 019 2 x-1 . 363 2 with corresponding coefficient of determination as 0. 991 9. It indicated that the image processing was simple,accurate and easy to apply to rapeseed.