猪三维点云体尺自动计算模型Pig Back Transformer
[目的/意义]为了提高体尺关键点定位准确率，猪三维点云体尺自动测量方法会采用点云分割，在各个分割后局部点云定位测量关键点，以减少点云之间相互干扰。然而点云分割网络通常需要消耗较大计算资源，且现有测量点定位效果仍有待提升空间。本研究旨在通过设计关键点生成网络从猪体点云中提取出各体尺测量所需关键点。在降低显存资源需求的同时提高测量关键点定位效果，提高体尺测量的效率和精度。[方法]针对猪三维表面点云进行体尺测量，提出了一种定位猪体尺关键点的模型Pig Back Transformer。模型分为两个模块，分别设计了两种改进的Transformer自注意力编码器，第一模块为全局关键点模块，首先设计了一种...
Saved in:
| Published in | 智慧农业（中英文） Vol. 6; no. 4; pp. 76 - 90 |
|---|---|
| Main Authors | , , , , , , |
| Format | Journal Article |
| Language | Chinese |
| Published |
中国农业科学院农业信息研究所
30.07.2024
国家生猪种业工程技术研究中心,广东广州 510642,中国 国家生猪种业工程技术研究中心,广东广州 510642,中国%华南农业大学 数学与信息学院,广东广州 510642,中国%国家生猪种业工程技术研究中心,广东广州 510642,中国 猪禽种业全国重点实验室,广东广州 510640,中国%华南农业大学 数学与信息学院,广东广州 510642,中国 华南农业大学 动物科学学院,广东广州 510642,中国 猪禽种业全国重点实验室,广东广州 510640,中国 |
| Subjects | |
| Online Access | Get full text |
| ISSN | 2096-8094 |
| DOI | 10.12133/j.smartag.SA202401023 |
Cover
| Abstract | [目的/意义]为了提高体尺关键点定位准确率，猪三维点云体尺自动测量方法会采用点云分割，在各个分割后局部点云定位测量关键点，以减少点云之间相互干扰。然而点云分割网络通常需要消耗较大计算资源，且现有测量点定位效果仍有待提升空间。本研究旨在通过设计关键点生成网络从猪体点云中提取出各体尺测量所需关键点。在降低显存资源需求的同时提高测量关键点定位效果，提高体尺测量的效率和精度。[方法]针对猪三维表面点云进行体尺测量，提出了一种定位猪体尺关键点的模型Pig Back Transformer。模型分为两个模块，分别设计了两种改进的Transformer自注意力编码器，第一模块为全局关键点模块，首先设计了一种猪背部边缘点提取算法用于获取边缘点，再使用edge encoder编码器以边缘点集合作为输入，edge encoder的edge attention中加入了边缘点和质点的偏移距离信息；第二模块为关键点生成模块，使用了back attention机制的back encoder，其中加入了与质心和第一模块生成的全局关键点的偏移量，并将偏移量与点云注意力通过按位max pooling操作结合，最后通过生成猪的体尺测量关键点和背脊走向点。最后设计了使用关键点和背脊走向点作为输入的体尺算法。[结果和讨论] 对比关键点和背脊走向点生成任务上Pig Back Transformer表现最佳，并对比体尺计算结果与人工测量结果，体长相对误差为0.63%，相对PointNet++、Point Transformer V2、Point Cloud Transforme、OctFormer PointTr等模型有较大提升。[结论]Pig Back Transformer能相对准确地生成猪体尺关键点，提高体尺测量数据准确度，并且通过点云特征定位体尺关键点节省了计算资源，为无接触牲畜体尺测量提供了新思路。 |
|---|---|
| AbstractList | S2; [目的/意义]为了提高体尺关键点定位准确率,猪三维点云体尺自动测量方法会采用点云分割,在各个分割后局部点云定位测量关键点,以减少点云之间相互干扰.然而点云分割网络通常需要消耗较大计算资源,且现有测量点定位效果仍有待提升空间.本研究旨在通过设计关键点生成网络从猪体点云中提取出各体尺测量所需关键点.在降低显存资源需求的同时提高测量关键点定位效果,提高体尺测量的效率和精度.[方法]针对猪三维表面点云进行体尺测量,提出了一种定位猪体尺关键点的模型Pig Back Transformer.模型分为两个模块,分别设计了两种改进的Transformer自注意力编码器,第一模块为全局关键点模块,首先设计了一种猪背部边缘点提取算法用于获取边缘点,再使用edge encoder编码器以边缘点集合作为输入,edge encoder的edge attention中加入了边缘点和质点的偏移距离信息;第二模块为关键点生成模块,使用了back attention机制的back encoder,其中加入了与质心和第一模块生成的全局关键点的偏移量,并将偏移量与点云注意力通过按位max pooling操作结合,最后通过生成猪的体尺测量关键点和背脊走向点.最后设计了使用关键点和背脊走向点作为输入的体尺算法.[结果和讨论]对比关键点和背脊走向点生成任务上Pig Back Transformer表现最佳,并对比体尺计算结果与人工测量结果,体长相对误差为0.63%,相对PointNet++、Point Transformer V2、Point Cloud Transforme、OctFormer PointTr等模型有较大提升.[结论]Pig Back Transformer能相对准确地生成猪体尺关键点,提高体尺测量数据准确度,并且通过点云特征定位体尺关键点节省了计算资源,为无接触牲畜体尺测量提供了新思路. [目的/意义]为了提高体尺关键点定位准确率，猪三维点云体尺自动测量方法会采用点云分割，在各个分割后局部点云定位测量关键点，以减少点云之间相互干扰。然而点云分割网络通常需要消耗较大计算资源，且现有测量点定位效果仍有待提升空间。本研究旨在通过设计关键点生成网络从猪体点云中提取出各体尺测量所需关键点。在降低显存资源需求的同时提高测量关键点定位效果，提高体尺测量的效率和精度。[方法]针对猪三维表面点云进行体尺测量，提出了一种定位猪体尺关键点的模型Pig Back Transformer。模型分为两个模块，分别设计了两种改进的Transformer自注意力编码器，第一模块为全局关键点模块，首先设计了一种猪背部边缘点提取算法用于获取边缘点，再使用edge encoder编码器以边缘点集合作为输入，edge encoder的edge attention中加入了边缘点和质点的偏移距离信息；第二模块为关键点生成模块，使用了back attention机制的back encoder，其中加入了与质心和第一模块生成的全局关键点的偏移量，并将偏移量与点云注意力通过按位max pooling操作结合，最后通过生成猪的体尺测量关键点和背脊走向点。最后设计了使用关键点和背脊走向点作为输入的体尺算法。[结果和讨论] 对比关键点和背脊走向点生成任务上Pig Back Transformer表现最佳，并对比体尺计算结果与人工测量结果，体长相对误差为0.63%，相对PointNet++、Point Transformer V2、Point Cloud Transforme、OctFormer PointTr等模型有较大提升。[结论]Pig Back Transformer能相对准确地生成猪体尺关键点，提高体尺测量数据准确度，并且通过点云特征定位体尺关键点节省了计算资源，为无接触牲畜体尺测量提供了新思路。 |
| Abstract_FL | [Objective]Nowadays most no contact body size measurement studies are based on point cloud segmentation method,they use a trained point cloud segmentation neural network to segment point cloud of pigs,then locate measurement points based on them.But point cloud segmentation neural network always need a larger graphics processing unit(GPU)memory,moreover,the result of the measurement key point still has room of improvement.This study aims to design a key point generating neural network to extract mea-surement key points from pig's point cloud.Reducing the GPU memory usage and improve the result of measurement points at the same time,improve both the efficiency and accuracy of the body size measurement.
[Methods]A neural network model was proposed using improved Transformer attention mechanic called Pig Back Transformer for generating key points and back orientation points which were related to pig body dimensions.In the first part of the network,it was in-troduced an embedding structure for initial feature extraction and a Transformer encoder structure with edge attention which was a self-attention mechanic improved from Transformer's encoder.The embedding structure using two shared multilayer perceptron(MLP)and a distance embedding algorithm,it takes a set of points from the edge of pig back's point cloud as input and then extract in-formation from the edge points set.In the encoder part,information about the offset distances between edge points and mass point which were their feature that extracted by the embedding structure mentioned before incorporated.Additionally,an extraction algo-rithm for back edge point was designed for extracting edge points to generate the input of the neural network model.In the second part of the network,it was proposed a Transformer encoder with improved self-attention called back attention.In the design of back atten-tion,it also had an embedding structure before the encoder structure,this embedding structure extracted features from offset values,these offset values were calculated by the points which are none-edge and down sampled by farthest point sampling(FPS)to both the relative centroid point and model generated global key point from the first part that introduced before.Then these offset values were processed with max pooling with attention generated by the extracted features of the points'axis to extract more information that the original Transformer encoder couldn't extract with the same number of parameters.The output part of the model was designed to gen-erate a set of offsets of the key points and points for back direction fitting,than add the set offset to the global key point to get points for pig body measurements.At last,it was introduced the methods for calculating body dimensions which were length,height,shoulder width,abdomen width,hip width,chest circumference and abdomen circumference using key points and back direction fitting points.
[Results and Discussions]In the task of generating key points and points for back direction fitting,the improved Pig Back Transformer performed the best in the accuracy wise in the models tested with the same size of parameters,and the back orientation points generat-ed by the model were evenly distributed which was a good preparation for a better body length calculation.A melting test for edge de-tection part with two attention mechanic and edge trim method both introduced above had being done,when the edge detection and the attention mechanic got cut off,the result had been highly impact,it made the model couldn't perform as well as before,when the edge trim method of preprocessing part had been cut off,there's a moderate impact on the trained model,but it made the loss of the model more inconsistence while training than before.When comparing the body measurement algorithm with human handy results,the relative error in length was 0.63%,which was an improvement compared to other models.On the other hand,the relative error of shoulder width,abdomen width and hip width had edged other models a little but there was no significant improvement so the perfor-mance of these measurement accuracy could be considered negligible,the relative error of chest circumference and abdomen circum-ference were a little bit behind by the other methods existed,it's because the calculate method of circumferences were not complicated enough to cover the edge case in the dataset which were those point cloud that have big holes in the bottom of abdomen and chest,it impacted the result a lot.
[Conclusions]The improved Pig Back Transformer demonstrates higher accuracy in generating key points and is more resource-effi-cient,enabling the calculation of more accurate pig body measurements.And provides a new perspective for non-contact livestock body size measurements. |
| Author | 吴珍芳 蔡更元 尹令 王宇啸 石源源 陈招达 张素敏 |
| AuthorAffiliation | 国家生猪种业工程技术研究中心,广东广州 510642,中国%华南农业大学 数学与信息学院,广东广州 510642,中国%国家生猪种业工程技术研究中心,广东广州 510642,中国;华南农业大学 动物科学学院,广东广州 510642,中国;猪禽种业全国重点实验室,广东广州 510640,中国%华南农业大学 数学与信息学院,广东广州 510642,中国;国家生猪种业工程技术研究中心,广东广州 510642,中国;猪禽种业全国重点实验室,广东广州 510640,中国 |
| AuthorAffiliation_xml | – name: 国家生猪种业工程技术研究中心,广东广州 510642,中国%华南农业大学 数学与信息学院,广东广州 510642,中国%国家生猪种业工程技术研究中心,广东广州 510642,中国;华南农业大学 动物科学学院,广东广州 510642,中国;猪禽种业全国重点实验室,广东广州 510640,中国%华南农业大学 数学与信息学院,广东广州 510642,中国;国家生猪种业工程技术研究中心,广东广州 510642,中国;猪禽种业全国重点实验室,广东广州 510640,中国 |
| Author_FL | WANG Yuxiao ZHANG Sumin CAI Gengyuan SHI Yuanyuan CHEN Zhaoda YIN Ling WU Zhenfang |
| Author_FL_xml | – sequence: 1 fullname: WANG Yuxiao – sequence: 2 fullname: SHI Yuanyuan – sequence: 3 fullname: CHEN Zhaoda – sequence: 4 fullname: WU Zhenfang – sequence: 5 fullname: CAI Gengyuan – sequence: 6 fullname: ZHANG Sumin – sequence: 7 fullname: YIN Ling |
| Author_xml | – sequence: 1 fullname: 王宇啸 organization: 国家生猪种业工程技术研究中心,广东广州510642,中国 – sequence: 2 fullname: 石源源 organization: 华南农业大学数学与信息学院,广东广州510642,中国 – sequence: 3 fullname: 陈招达 organization: 华南农业大学数学与信息学院,广东广州510642,中国 – sequence: 4 fullname: 吴珍芳 organization: 华南农业大学数学与信息学院,广东广州510642,中国 – sequence: 5 fullname: 蔡更元 organization: 华南农业大学数学与信息学院,广东广州510642,中国 – sequence: 6 fullname: 张素敏 organization: 华南农业大学数学与信息学院,广东广州510642,中国 – sequence: 7 fullname: 尹令 organization: 华南农业大学数学与信息学院,广东广州510642,中国 |
| BookMark | eNotjz1Lw0Acxm-oYK39ClIEx9T_3eVyCU61qBWKClakXcIlTaLRXiCniI4u4ssiDm5Ct9KhODhoxW-TmH4MQ-P08MCP52UJlWQkPYRWMNQxwZSuh3U1EPGFCOqHDQJEBwyEllCZgGVoJlj6IqoqFQIAsTAhnJXRRvY0Tj7vs--P7PYrmT4nPy_p-3R2N04fRrPJMJu8_o6G6dvjwWlQ2xTuWa0TC6n8KB548TJa8MW58qr_WkFH21udZktr7-_sNhttzcWEc83ELvN8x2TUZGB4TLdyhw0hCOWW6FPK8mkUqO4AdRn3gDvYcXTTtMARvkFpBa0VuVdC-kIGdhhdxjJvtG9O5PX8qA7Ac2614FyhlC2V6tu91l631z2eMwX0B3VzZJI |
| ClassificationCodes | S2 |
| ContentType | Journal Article |
| Copyright | Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| Copyright_xml | – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| DBID | NSCOK 2B. 4A8 92I 93N PSX TCJ |
| DOI | 10.12133/j.smartag.SA202401023 |
| DatabaseName | 国家哲学社会科学文献中心 Wanfang Data Journals - Hong Kong WANFANG Data Centre Wanfang Data Journals 万方数据期刊 - 香港版 China Online Journals (COJ) China Online Journals (COJ) |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| DocumentTitle_FL | Pig Back Transformer:Automatic 3D Pig Body Measurement Model |
| EndPage | 90 |
| ExternalDocumentID | zhny202404007 ZHNYZYW2024004007 |
| GroupedDBID | ABDBF ACUHS ALMA_UNASSIGNED_HOLDINGS EDH GROUPED_DOAJ NSCOK TCJ TGD U1G U5N 2B. 4A8 92I 93N PSX |
| ID | FETCH-LOGICAL-c1277-81c5efb8538506e549efb16aa2379ad3358093034b03c57e07b1bb48890baf633 |
| ISSN | 2096-8094 |
| IngestDate | Thu May 29 04:06:09 EDT 2025 Mon Feb 17 13:28:39 EST 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 4 |
| Keywords | 三维点云 自注意力机制 Transformer Back 体尺自动测量 深度相机 Pig 测量关键点定位 key point positioning body size automic measurement 3D point cloud self-atten-tion mechanism Pig Back Transformer depth camera |
| Language | Chinese |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c1277-81c5efb8538506e549efb16aa2379ad3358093034b03c57e07b1bb48890baf633 |
| OpenAccessLink | http://dx.doi.org/10.12133/j.smartag.SA202401023 |
| PageCount | 15 |
| ParticipantIDs | wanfang_journals_zhny202404007 cass_nssd_ZHNYZYW2024004007 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-07-30 |
| PublicationDateYYYYMMDD | 2024-07-30 |
| PublicationDate_xml | – month: 07 year: 2024 text: 2024-07-30 day: 30 |
| PublicationDecade | 2020 |
| PublicationTitle | 智慧农业（中英文） |
| PublicationTitle_FL | Smart Agriculture |
| PublicationYear | 2024 |
| Publisher | 中国农业科学院农业信息研究所 国家生猪种业工程技术研究中心,广东广州 510642,中国 国家生猪种业工程技术研究中心,广东广州 510642,中国%华南农业大学 数学与信息学院,广东广州 510642,中国%国家生猪种业工程技术研究中心,广东广州 510642,中国 猪禽种业全国重点实验室,广东广州 510640,中国%华南农业大学 数学与信息学院,广东广州 510642,中国 华南农业大学 动物科学学院,广东广州 510642,中国 猪禽种业全国重点实验室,广东广州 510640,中国 |
| Publisher_xml | – name: 中国农业科学院农业信息研究所 – name: 华南农业大学 动物科学学院,广东广州 510642,中国 – name: 国家生猪种业工程技术研究中心,广东广州 510642,中国%华南农业大学 数学与信息学院,广东广州 510642,中国%国家生猪种业工程技术研究中心,广东广州 510642,中国 – name: 国家生猪种业工程技术研究中心,广东广州 510642,中国 – name: 猪禽种业全国重点实验室,广东广州 510640,中国 – name: 猪禽种业全国重点实验室,广东广州 510640,中国%华南农业大学 数学与信息学院,广东广州 510642,中国 |
| SSID | ssj0002912275 ssib057733647 ssib051372935 ssib042363126 |
| Score | 2.3918827 |
| Snippet | [目的/意义]为了提高体尺关键点定位准确率，猪三维点云体尺自动测量方法会采用点云分割，在各个分割后局部点云定位测量关键点，以减少点云之间相互干扰。然而点云分割网络... S2; [目的/意义]为了提高体尺关键点定位准确率,猪三维点云体尺自动测量方法会采用点云分割,在各个分割后局部点云定位测量关键点,以减少点云之间相互干扰.然而点云分割网络... |
| SourceID | wanfang cass |
| SourceType | Aggregation Database |
| StartPage | 76 |
| Title | 猪三维点云体尺自动计算模型Pig Back Transformer |
| URI | https://www.ncpssd.cn/Literature/articleinfo?id=ZHNYZYW2024004007&type=eJournalArticle&typename=中文期刊文章&nav=1&langType=1&pageUrl=https%253A%252F%252Fwww.ncpssd.org%252Fjournal%252Fdetails%253Fgch%253D211192%2526nav%253D1%2526langType%253D2 https://d.wanfangdata.com.cn/periodical/zhny202404007 |
| Volume | 6 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals issn: 2096-8094 databaseCode: DOA dateStart: 20190101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.doaj.org/ omitProxy: true ssIdentifier: ssj0002912275 providerName: Directory of Open Access Journals – providerCode: PRVEBS databaseName: EBSCOhost Academic Search Ultimate issn: 2096-8094 databaseCode: ABDBF dateStart: 20210901 customDbUrl: https://search.ebscohost.com/login.aspx?authtype=ip,shib&custid=s3936755&profile=ehost&defaultdb=asn isFulltext: true dateEnd: 99991231 titleUrlDefault: https://search.ebscohost.com/direct.asp?db=asn omitProxy: true ssIdentifier: ssj0002912275 providerName: EBSCOhost – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources issn: 2096-8094 databaseCode: M~E dateStart: 20200101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://road.issn.org omitProxy: true ssIdentifier: ssib057733647 providerName: ISSN International Centre |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnR1NaxQxNLT14kUUFWu17MF3kq2Zj2QSPGV2ZymCRbDFtpdlPlsVV3BXxB69iB8X8eBN6K30UDx40Ir_puv2Z_hedroz0EVUWIbMy3tv3sck7yWbZBi7wTNP-AUvmpmvdNNXrtNUMnWbfiFFnGHCX9hzuu-uyOU1_866WJ-ZfVHfXTJIltKdqftK_serCEO_0i7Zf_DshCkCsIz-xSt6GK9_5WOIAlAtMAYiH0IFShMkDCH0bZULobZVBrRjC23QHkQCQk7ACEkCSy5AGTCKICYC4xA5FnQAkSQ4QQToCFR47-HWzTBOH4_PRaect1zg--hEKAlaW_YSlAATWPYIbJViaqzqQNgCpUqIaVtZUHDHkkuSq8SZzFRYlUgC4ofCKctYozKqjqKRzCMupDWfFCoUTeLRoyXx0iE9OozwV6EIoiyN2AE1ls4g3_okievb2Vc-ea1rygjiG7anaR6QRbS1J2IaWQpk3GnIWOhY46OsDpiOJeegrWgGrSxtFerDqy7dxQEj5gTjzzqfxB9Za2Z-LZYEspaVjM10Kt65Dk3IY8DrP8G3MN5aum9IfToo0Ksi_GTd5ebyysbmxgOLQr13MMvOuBgOeW0uAjtiTLGl51R5uXDoz90qrxYBHaJZjkMp5XG147r2UOuJhuVufBLw1lTxMAdKcXxoN8v1iri3VcvrVs-zc-WArGHGresCm9nZvshuj97vH317M_rxdfTq-9Hhh6OfH4dfDo9f7w_f7h0f7I4OPv3a2x1-fodtoUFtoVFrC5fYWidabS03y8-MNFOHFjAoJxV5kWDeSqc35sLXeOfIOHa9QMeZRwsFNGZ6fsK9VAQ5DxInSTDwaZ7EhfS8y2yu97SXX2ENlSEgyRIeaETIY8wkfK6TPNPcS3JRzLMF0rnb6_ez7ilnzLPF0hTdspPpd3e2ey8tBiFc_SP5AjtbvfrX2Nzg2fP8OmbMg2TRevc3BwCMQg |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E7%8C%AA%E4%B8%89%E7%BB%B4%E7%82%B9%E4%BA%91%E4%BD%93%E5%B0%BA%E8%87%AA%E5%8A%A8%E8%AE%A1%E7%AE%97%E6%A8%A1%E5%9E%8BPig+Back+Transformer&rft.jtitle=%E6%99%BA%E6%85%A7%E5%86%9C%E4%B8%9A%EF%BC%88%E4%B8%AD%E8%8B%B1%E6%96%87%EF%BC%89&rft.au=%E7%8E%8B%E5%AE%87%E5%95%B8&rft.au=%E7%9F%B3%E6%BA%90%E6%BA%90&rft.au=%E9%99%88%E6%8B%9B%E8%BE%BE&rft.au=%E5%90%B4%E7%8F%8D%E8%8A%B3&rft.date=2024-07-30&rft.pub=%E4%B8%AD%E5%9B%BD%E5%86%9C%E4%B8%9A%E7%A7%91%E5%AD%A6%E9%99%A2%E5%86%9C%E4%B8%9A%E4%BF%A1%E6%81%AF%E7%A0%94%E7%A9%B6%E6%89%80&rft.issn=2096-8094&rft.volume=6&rft.issue=4&rft.spage=76&rft.epage=90&rft_id=info:doi/10.12133%2Fj.smartag.SA202401023&rft.externalDocID=ZHNYZYW2024004007 |
| thumbnail_s | http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fzhny%2Fzhny.jpg |