基于机器视觉的梭子蟹质量估计方法研究
PDF下载 (598)张 超 1,徐建瑜 1*,王文静 2.基于机器视觉的梭子蟹质量估计方法研究[J].宁波大学学报(理工版),2014,27(02):49-51.DOI:
ZHANG Chao 1,XU Jian-yu 1*,WANG Wen-jing 2.Machine Vision Based Weight Estimation of Swimming Crab[J].Journal of Ningbo University(Natural Science & Engineering Edition),2014,27(02):49-51.DOI:
| Title: | Machine Vision Based Weight Estimation of Swimming Crab |
| 作者: | 张 超 1, 徐建瑜 1*, 王文静 2 |
| Author(s): | ZHANG Chao 1, XU Jian-yu 1*, WANG Wen-jing 2 |
| 关键词: | 梭子蟹; 质量估计; 图像处理技术; 数据拟合 |
| Keywords: | swimming crab; weight estimation; image processing; data fitting |
| 分类号: | TP183 |
| 文献标识码: | A |
| 摘要: | 传统梭子蟹养殖定期质量估计的方法是捕捞部分梭子蟹进行称重, 会对其造成物理和应激伤害, 且检测过程中受到人为因素干扰, 因此利用机器视觉技术对梭子蟹的质量进行估计. 通过摄像头拍摄不同生长阶段的梭子蟹图像, 对图像进行模板校正及图像分割, 提取梭子蟹面积特征参数. 利用最小二乘法对面积和质量进行拟合, 其中二次多项式相关性最好可达到0.9220, 测试平均相对误差6.40%, 证明该方法可以满足梭子蟹质量估计的要求. |
| Abstract: | The traditional means of weight estimation for swimming crabs is catching some samples to weigh, which may raise some unwanted issues such as safety problems, high labor cost, time-consuming, human subjectivity and low efficiency. In this paper, the machine vision technology is applied to facilitate the weighing process. First, the image and the weight data of a swimming crab at different growth stages are collected. Then the image is digitally processed for correction and segmentation through which the area of swimming crab can thus be identified. Data fitting between area and weight is conducted using the least squares technique. It is found that the best fitting can be obtained using quadratic polynomial with correlation coefficient being 0.9220 and average relative being 6.40%, suggesting the validation of the proposed method for weight estimation of swimming crabs. |
| 参考文献 /References: | [1]. 计算机视觉技术在水产养殖中的应用[J]. 浙江海洋学院学报, 2008, 27(4):439-443. [2].徐建瑜, 崔绍荣, 苗香雯, 等. 计算机视觉技术在水产养殖中应用与展望[J]. 农业工程学报, 2005, 21(8):174-178. [3].刘伟, 谭鹤群, 黄丹, 等. 白鲢质量与截面积沿体长方向分布模型[J]. 农业工程学报, 2012, 28(12):288-292. [4].王文静, 徐建瑜, 吕志敏, 等. 基于机器视觉水下鲆鲽鱼类质量估计[J]. 农业工程学报, 2012, 28(16):153-157. [5].Mathiassen J R, Misimi E, Toldnes B, et al. High-speed weight estimation of whole Herrin (Clupea harengus) using 3D machine vision[J]. Journal of Food Science, 2011, 76(6):1-7. [6].肖刚, 应晓芳, 高飞, 等. 基于领域灰度差值的二维Ostu分割方法研究[J]. 计算机应用研究, 2009, 26(4): 1544-1547 [7].赵晖, 林成龙, 唐朝京. 基于峰值聚类趋势检验含噪声图像快速阈值分割法[J]. 信号处理, 2009, 25(11):1666-1674. [8].张红涛, 毛罕平, 邱道尹. 储粮害虫图像识别中的特征提取[J]. 农业工程学报, 2009, 25(2):126-130. [9].Diao J, Lei X D, Hong L X, et al. Single leaf area estimation models based on leaf weight of eucalyptus[J]. Journal of Forestry Research, 2010, 21(1):73-76. [10].Gümü? B, Balaban M O. Prediction of the weight of aquacultured rainbow trout (Oncorhynchus mykiss) by image analysis[J]. Journal of Aquatic Food Product Technology, 2010, 19(3/4):227-237. |
| 备注/Memo: | 收稿日期: 2013-07-20. 宁波大学学报（理工版）网址 http://journallg.nbu.edu.cn/ 基金项目: 浙江省重大科技攻关专项计划项目（2011C11049）; 宁波市自然科学基金（2013A610158）. 第一作者: 张 超（1989－）, 男, 湖北黄冈人, 在读硕士研究生, 主要研究方向: 计算机视觉与图像处理. E-mail: 352498686@qq.com *通信作者: 徐建瑜（1973－）, 女, 黑龙江牡丹江人, 博士/副教授, 主要研究方向: 生物图像处理. E-mail: xujianyu@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
