ass日本风韵熟妇pics男人扒开女人屁屁桶到爽|扒开胸露出奶头亲吻视频|邻居少妇的诱惑|人人妻在线播放|日日摸夜夜摸狠狠摸婷婷|制服 丝袜 人妻|激情熟妇中文字幕|看黄色欧美特一级|日本av人妻系列|高潮对白av,丰满岳妇乱熟妇之荡,日本丰满熟妇乱又伦,日韩欧美一区二区三区在线

基于局部點(diǎn)云的蘋(píng)果外形指標(biāo)估測(cè)方法
CSTR:
作者:
作者單位:

作者簡(jiǎn)介:

通訊作者:

中圖分類(lèi)號(hào):

基金項(xiàng)目:

國(guó)家自然科學(xué)基金項(xiàng)目(31601545)和中央高校基本科研業(yè)務(wù)費(fèi)專項(xiàng)資金項(xiàng)目(KJQN201732)


Apple Shape Index Estimation Method Based on Local Point Cloud
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 圖/表
  • |
  • 訪問(wèn)統(tǒng)計(jì)
  • |
  • 參考文獻(xiàn)
  • |
  • 相似文獻(xiàn)
  • |
  • 引證文獻(xiàn)
  • |
  • 資源附件
  • |
  • 文章評(píng)論
    摘要:

    為了獲取果實(shí)生長(zhǎng)期的外形參數(shù)指標(biāo),監(jiān)控果實(shí)發(fā)育狀況,提出了一種基于局部點(diǎn)云的蘋(píng)果外形指標(biāo)估測(cè)方法。該方法可以通過(guò)局部點(diǎn)云數(shù)據(jù)估測(cè)蘋(píng)果的體積、高度、直徑等外形指標(biāo)參數(shù)。利用橢球曲面方程構(gòu)建蘋(píng)果幾何模型,并計(jì)算蘋(píng)果幾何模型的高度、直徑、體積。使用Kinect V2相機(jī)從任意角度獲取點(diǎn)云數(shù)據(jù),采用直通濾波法去除點(diǎn)云數(shù)據(jù)的背景,用包圍盒算法精簡(jiǎn)點(diǎn)云得到蘋(píng)果局部點(diǎn)云數(shù)據(jù)后,采用粒子群算法將蘋(píng)果局部點(diǎn)云數(shù)據(jù)與蘋(píng)果模型進(jìn)行空間匹配,并用遺傳算法求解蘋(píng)果最優(yōu)匹配模型的參數(shù),利用蘋(píng)果最優(yōu)匹配模型參數(shù)估測(cè)與其匹配的真實(shí)蘋(píng)果的外形指標(biāo)。實(shí)驗(yàn)采集了250個(gè)蘋(píng)果頂部、側(cè)面和底部的局部點(diǎn)云數(shù)據(jù),使用本文方法分別估測(cè)了250個(gè)蘋(píng)果在3個(gè)角度下的外形指標(biāo),并對(duì)估測(cè)值與真實(shí)值進(jìn)行線性回歸分析,各個(gè)指標(biāo)的線性回歸擬合度R2均高于0.7。其中,側(cè)面拍攝時(shí)擬合效果最好,R2最高為0.948。在各個(gè)角度下蘋(píng)果體積估測(cè)的平均誤差不大于16.16mL,高度估測(cè)的平均誤差不大于2.92mm,直徑估測(cè)的平均誤差不大于2.35mm,估測(cè)結(jié)果的平均誤差較小,在允許誤差范圍內(nèi)。實(shí)驗(yàn)結(jié)果表明,基于局部點(diǎn)云的蘋(píng)果外形指標(biāo)估測(cè)方法具有較強(qiáng)的實(shí)用性。

    Abstract:

    In order to obtain the shape parameters of growing fruit and monitor the fruit development status, an apple point index estimation method based on local point cloud was proposed. The method could estimate the shape index parameters such as volume, height and diameter of apple through apple local point cloud data. Firstly, the geometric model of apple was constructed by using the method of ellipsoidal surface equation, and the height, diameter and volume of apple geometric model were calculated. Kinect V2 was used to get point cloud data from any angle. Secondly, the passthrough filtering method was used to remove the background of point cloud data and the bounding box reduction algorithm was used to streamline the point cloud, and then the apple’s local point cloud was obtained. After that, the genetic algorithm was used to solve the optimal apple geometric model parameters. Finally, the height, diameter and volume of apple optimal matching model were used to estimate the shape index parameters of matching apple. The experiment collected local point cloud data of 250 apples at three different angles, namely the top, side and bottom of apple. Using this method, the shape indicators of 250 apples were estimated under these three angles. A linear regression method was used to analyze the linear correlation between the estimated value and the true value. The linear regression fit of each indicator was higher than 0.7. Among them, when the angle was the side of the apple, the linear regression fitting effect was the best, and the R2 was up to 0.948. And the average error of the apple volume estimation results under angles was no more than 16.16mL, the average error of the height estimation result was no more than 2.92mm, the average error of the diameter estimation result was no more than 2.35mm, and the average error was within the allowable error range. The experimental results showed that the method was stable and practical. 

    參考文獻(xiàn)
    相似文獻(xiàn)
    引證文獻(xiàn)
引用本文

王浩云,閆茹琪,周小莉,馬仕航,胡皓翔,徐煥良.基于局部點(diǎn)云的蘋(píng)果外形指標(biāo)估測(cè)方法[J].農(nóng)業(yè)機(jī)械學(xué)報(bào),2019,50(5):205-213. WANG Haoyun, YAN Ruqi, ZHOU Xiaoli, MA Shihang, HU Haoxiang, XU Huanliang. Apple Shape Index Estimation Method Based on Local Point Cloud[J]. Transactions of the Chinese Society for Agricultural Machinery,2019,50(5):205-213.

復(fù)制
相關(guān)視頻

分享
文章指標(biāo)
  • 點(diǎn)擊次數(shù):
  • 下載次數(shù):
  • HTML閱讀次數(shù):
  • 引用次數(shù):
歷史
  • 收稿日期:2019-03-07
  • 最后修改日期:
  • 錄用日期:
  • 在線發(fā)布日期: 2019-05-10
  • 出版日期: 2019-05-10
文章二維碼