基于三维线框模型的类球体农产品体积和表面积测量

发布时间：2018-06-25 16:36

本文选题：类球体农产品 + 体积　；参考：《农业机械学报》2016年07期

【摘要】：为了快速无损测量不规则类球体形状农产品的体积和表面积,提出了基于三维线框模型的图像处理方法。即以Android系统为工作平台,利用Java结合OpenCV函数库,以三维线框模型为理论指导编写包括图像采集、图像分割、体积和表面积计算的应用软件。用3种不同直径的标准球体在6个拍摄距离下进行测量,所获测量值与理论值的误差小于0.3%,表明摄像头自身畸变小,在图像能完整呈现被测物的拍摄距离,被测物轮廓的大小对测量精度的影响在可接受的误差范围之内。用所提方法、排水法和扫描图像处理法对不规则形状的柑橘、苹果、梨各10组进行体积和表面积测量,结果显示3种测量方法的相关系数大于0.98,表明该方法可较精确测量类球体农产品的体积和表面积。
[Abstract]:An image processing method based on 3D wire-frame model was proposed to measure the volume and surface area of agricultural products with irregular spherical shape quickly. Using Android system as the working platform, the application software including image acquisition, image segmentation, volume and surface area calculation is developed under the guidance of three-dimensional wire-frame model and using Android combined with OpenCV function library. Three standard spheres of different diameters were used to measure at 6 shooting distances. The error between the measured values and the theoretical values was less than 0.3, which indicated that the camera itself had less distortion and could show the shooting distance of the measured objects completely in the image. The effect of the size of the measured object profile on the measurement accuracy is within the acceptable error range. The volume and surface area of 10 groups of irregularly shaped citrus, apple and pear were measured by the proposed method, drainage method and scanning image processing method. The results show that the correlation coefficient of the three measuring methods is greater than 0.98, which indicates that this method can accurately measure the volume and surface area of spherical agricultural products.
【作者单位】：深圳信息职业技术学院机电学院;郑州轻工业学院计算机与通信工程学院;浙江大学生物系统工程与食品科学学院;
【基金】：国家自然科学基金项目(61405175) 深圳市战略新兴产业发展专项(JCYJ20150417094158023)
【分类号】：TP391.41;S126

【参考文献】