|
| 基于机器视觉的小龙虾苗种质量估计研究 |
| Research on the Weight Estimation of Crayfish Fry Based on Machine Vision |
| |
| DOI：10.16018/j.cnki.cn32-1650/n.202501009 |
| 中文关键词: 机器视觉 小龙虾苗种 图像处理 质量估计 数据拟合 |
| 英文关键词: machine vision crayfish fry image processing weight estimation data fitting |
| 基金项目: |
|
| 摘要点击次数: 112 |
| 全文下载次数: 93 |
| 中文摘要: |
| 针对传统小龙虾苗种分拣方法会对苗种造成物理和应激伤害，且小龙虾苗种耐受性较低的问题，本文提出一种基于机器视觉的小龙虾苗种质量估计方法，以实现无接触式优质苗种质量估计。通过工业相机拍摄不同规格的小龙虾苗种图像，首先对小龙虾苗种图像进行预处理，然后采用K-means聚类算法进行图像分割，最后运用形态学运算和泛洪算法处理提取小龙虾苗种轮廓，并使用像素统计法提取小龙虾苗种像素数量特征参数。利用最小二乘法对像素数量和质量进行数据拟合，其中二次多项式拟合相关性最好可达到0. 954 0，测试平均相对误差4. 60%，实验证明该方法可以满足小龙虾苗种质量估计的要求。 |
| 英文摘要: |
| In order to solve the problems that the traditional sorting method of crayfish fry will cause physical and stress damage to the fry, and the tolerance of crayfish fry is low, this paper proposes a method of crayfish fry weight estimation based on machine vision to realize non-contact weight estimation of crayfish fry. The crayfish fry images with different specifications were photographed by industrial cameras. Firstly, the crayfish fry images were preprocessed, and then the images were segmented by K-means clustering algorithm. Finally, the outline of crayfish fry was extracted by morphological operation and flooding algorithm, and the pixel number characteristic parameters of crayfish fry were extracted by pixel statistics. The least squares method is used to perform data fitting on the number of pixels and weight. Among them, the correlation of quadratic polynomial fitting is the best, which can reach 0. 954 0, and the average relative error of the test is 4. 60%. The experiment proves that this method can meet the requirements for the weight estimation of crayfish fry. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|