化石講場-Fossils Board 
» 遊客:  註冊 | 登錄 | 會員 | 幫助

RSS 訂閱當前論壇  


上一主題 下一主題
     
標題: Computer vision cracks the leaf code  
 
Wong
[藻類] 新手上路
Rank: 1



UID 39658
精華 0
積分 14
帖子 10
閱讀權限 10
註冊 2016-4-11
來自 北京
狀態 離線
[廣告]:
Computer vision cracks the leaf code

Go to PNAS Homepage
>  Current Issue
> vol. 113 no. 12
> Peter Wilf,  3305–3310, doi: 10.1073/pnas.1524473113

Computer vision cracks the leaf code

Peter Wilfa,1,
Shengping Zhangb,c,1,
Sharat Chikkerurd,
Stefan A. Littlea,e,
Scott L. Wingf, and
Thomas Serreb,1

Author Affiliations

Edited by Andrew H. Knoll, Harvard University, Cambridge, MA, and approved February 1, 2016 (received for review December 14, 2015)

Significance

The botanical value of angiosperm leaf shape and venation (“leaf architecture”) is well known, but the astounding complexity and variation of leaves have thwarted efforts to access this underused resource. This challenge is central for paleobotany because most angiosperm fossils are isolated, unidentified leaves. We here demonstrate that a computer vision algorithm trained on several thousand images of diverse cleared leaves successfully learns leaf-architectural features, then categorizes novel specimens into natural botanical groups above the species level. The system also produces heat maps to display the locations of numerous novel, informative leaf characters in a visually intuitive way. With assistance from computer vision, the systematic and paleobotanical value of leaves is ready to increase significantly.


Abstract

Understanding the extremely variable, complex shape and venation characters of angiosperm leaves is one of the most challenging problems in botany. Machine learning offers opportunities to analyze large numbers of specimens, to discover novel leaf features of angiosperm clades that may have phylogenetic significance, and to use those characters to classify unknowns. Previous computer vision approaches have primarily focused on leaf identification at the species level. It remains an open question whether learning and classification are possible among major evolutionary groups such as families and orders, which usually contain hundreds to thousands of species each and exhibit many times the foliar variation of individual species. Here, we tested whether a computer vision algorithm could use a database of 7,597 leaf images from 2,001 genera to learn features of botanical families and orders, then classify novel images. The images are of cleared leaves, specimens that are chemically bleached, then stained to reveal venation. Machine learning was used to learn a codebook of visual elements representing leaf shape and venation patterns. The resulting automated system learned to classify images into families and orders with a success rate many times greater than chance. Of direct botanical interest, the responses of diagnostic features can be visualized on leaf images as heat maps, which are likely to prompt recognition and evolutionary interpretation of a wealth of novel morphological characters. With assistance from computer vision, leaves are poised to make numerous new contributions to systematic and paleobotanical studies.
leaf architecture
leaf venation
computer vision
sparse coding


圖片附件: 未标题-1 拷贝.jpg (2016-4-17 08:56 AM, 170.54 K)





古植物是化石的歌!
2016-4-17 08:56 AM#1
查看資料  發短消息  頂部
 
oviraptor (偷蛋龍)
超級版主
Rank: 8Rank: 8


版主勳章   貢獻勳章   創意勳章   演講勳章   榮譽勳章   出席勳章  
UID 2
精華 4
積分 1960
帖子 1244
閱讀權限 150
註冊 2006-10-16
來自 香港
狀態 離線
[廣告]:
Thanks for sharing. A living plants?



(\\\\\) (\\\\\) (\\\\\) (/////) (/////) (/////)
2016-4-17 02:40 PM#2
查看資料  發短消息  頂部
 
Wong
[藻類] 新手上路
Rank: 1



UID 39658
精華 0
積分 14
帖子 10
閱讀權限 10
註冊 2016-4-11
來自 北京
狀態 離線
[廣告]:
计算机视觉破解叶子编码

    认识被子植物叶子多变而复杂的形状和脉式特征是植物学中最具挑战性的问题之一。机器学习提供了机会用来分析大量的标本,从而发现具有系统发育含义的被子植物支系的革新特征,据此来分类未知的标本。先前的计算机可视方法主要用于叶片种级水平的鉴定。机器是否可能学习和分类仍待商榷,特别是针对大的进化类群(科和目),通常包含成百上千的具有多次叶变异的种。本文中我们测试了一种计算机可视算法是否能利用一个由2001个属的7597幅叶图像数据库来学习植物学科和目的特征,进而分类新的图像。图像是通过化学漂洗处理过的透明叶标本,然后染色后显示了叶脉。机器学习可用来学习一种代表了叶形状和脉式的编码本。最终成功率远大于偶然性,自动化系统能学着把图像分类成科和目。其中最饶有植物学兴趣是,对叶图像上鉴别特征的反应能以热感应图的形式看见,它们可能促进对大量新形态特征的识别和演化解释。通过计算机视图的帮助,叶子对系统学和古植物学研究一定能做出很多新的贡献。

----------------------------------------------------------------------------


今年三月计算机和人下围棋是一个实践的例子:人工智能不可小觑!!!
阿尔法围棋(AlphaGo)是一款围棋人工智能程序,由位于英国伦敦的谷歌(Google)旗下DeepMind公司的戴维·西尔弗、艾佳·黄和戴密斯·哈萨比斯与他们的团队开发,这个程序利用“价值网络”去计算局面,用“策略网络”去选择下子。2015年10月阿尔法围棋以5:0完胜欧洲围棋冠军、职业二段选手樊麾;2016年3月对战世界围棋冠军、职业九段选手李世石,并以4:1的总比分获胜 。

[ 本帖最後由 Wong 於 2016-4-17 08:48 PM 編輯 ]




古植物是化石的歌!
2016-4-17 08:09 PM#3
查看資料  發短消息  頂部
 
Wong
[藻類] 新手上路
Rank: 1



UID 39658
精華 0
積分 14
帖子 10
閱讀權限 10
註冊 2016-4-11
來自 北京
狀態 離線
[廣告]:
回復 #2 oviraptor 的帖子

Here I grossly translated this English abstract into Chinese. The authors should make use of the cleared leaves of living angiosperms. The result will be promising and helpful for us to study fossil leaves. The first author Peter Wilf is a paleobotanist.



古植物是化石的歌!
2016-4-17 08:15 PM#4
查看資料  發短消息  頂部
 
oviraptor (偷蛋龍)
超級版主
Rank: 8Rank: 8


版主勳章   貢獻勳章   創意勳章   演講勳章   榮譽勳章   出席勳章  
UID 2
精華 4
積分 1960
帖子 1244
閱讀權限 150
註冊 2006-10-16
來自 香港
狀態 離線
[廣告]:
回復 #4 Wong 的帖子

Thanks



(\\\\\) (\\\\\) (\\\\\) (/////) (/////) (/////)
2016-5-4 09:14 PM#5
查看資料  發短消息  頂部
     


  可打印版本 | 推薦給朋友 | 訂閱主題 | 收藏主題  


 


{支持化石講場客戶廣告, 謝謝!}


{香港首個古生物化石網站 - 化石專賣/籌辦展覽/教育推廣/復原工作}
恐龍@化石工作室 Dinosaur @ Fossil Lab



本论坛支付平台由支付宝提供
携手打造安全诚信的交易社区   Powered by Discuz! 4.1.0  © 2001-2006 Comsenz Inc.
Processed in 0.014230 second(s), 7 queries

所有時間為 GMT+8, 現在時間是 2024-10-9 05:11 PM 清除 Cookies - 聯繫我們 - 恐龍@化石工作室 - Archiver - WAP