Rice Phenomic Network
Chairman : Pisuth Paiboonrat (khunpisuth AT gmail.com)
Members : Ye-Nu Wan (Session chair)
Takuji Kiura (Session Co-chair)
AG members
invited International network in specific fields
Objectives : Rice is a staple crop in Asia. The climate change phenomena has affected the regional rice production in active and passive situation. To understand crop performance in responsive to micro and macro-climatic variability is become a priority. Genotype and phenotype in interact with climatic condition needs a strong and commitment collaboration network. Thus to have a research back up for food security for the region.
Target Audience : Researchers in agricultural sciences, Researchers in climatic study, Researchers in embedded technologies, Students, Farmers organization leaders, Agricultural policy makers.
Expected Number of Participants : 30
Agenda : Ye-Nu Wan (Session chair), Takuji Kiura (Session Co-chair)

Field based phenotyping approaches by using time series images of paddy rice

Authors: Wei Guo and Seishi Ninomiya (U. Tokyo, Japan) [Remote]

Keywords:crop phenotypic traits, image analysis, machine learning


Rice is the most important staple crop in Asian countries and several core breeding programs are undertaken to meet the increasing demand and to improve quality. However, the traditional methods to investigate the crop phenotypic traits are labor intensive and time consuming (low-throughput phenotyping). In this study we proposed new high-throughput phenotyping approaches by analyzing paddy rice images taken under natural field conditions. We first proposed an illumination invariant vegetation segmentation tool DTSM (decision tree based segmentation model), which is proved can effective and efficient segment the crop regions from field taken images with different illuminate conditions. Then by using DTSM we evaluated the Leaf emergence pattern in initial growing stage, the growth pattern of crop canopy coverage from transplanting to maturing. The accuracy of the estimated canopy coverage rate was as good as 99%. Finally, we proposed a method that automatically detect the flowering parts of rice panicles and emulate diurnal flowering pattern during the flowering stage. The number of the detected flowering panicles was highly correlated with the visually counted number. Both of the proposed approaches are based on machine learning techniques and, once the tools are well trained by appropriate training data, they are simply applicable to huge amount of time series images without any model adjustment.

Rice mutant phenomics in the post-genomic era

Authors: Chang-Sheng Wang and Da-Gin Lin (N. Chung-Hsing University, Taiwan Agricultural Research Institute, Taiwan)


Rice genome has been completely sequenced we are now in the post- genomic era. The concept of Central Dogma has to expand to the New Central Dogma to cover the whole genome. Functional study of genome needs the linking of genotype and phenotype information. As the availability of genomic sequences is getting easy and cheap day by day, the tedious and time consuming phenotype investigation become the limitation of genetics and functional genomics study. Phenotyping of rice plant during growing period is a critical step to analyze gene function. In the past 15 years, we have developed two novel rice mutation pools with wide genetic diversity for the TNG67 (a japonica type) and the IR64 (an indica type) varieties, respectively. Mutants derived from the same variety share most genomic background and provide good materials for the functional study of rice genome. Genes and molecular markers can be developed from mutants by genetic approach with the integration of genomic sequences decoding by the next-generation sequencing (NGS) technology. Breeding in the post-genomic era could be more efficient if the high-through-put phenotyping, the phenomics, system can be established. - Slides

Morphological Analysis of Rice Seeds

Authors: Tzu-Yi Kuoa, Szu-Yu Chenb, Heng-An Linb, Chia-Lin Chungb, Yan-Fu Kuoa
aDepartment of Bio-industrial Mechatronics Engineering,
bDepartment of Plant Pathology and Microbiology,
National Taiwan University, Taiwan, R.O.C.


Rice is a staple food worldwide. This study quantitatively analyzed the morphological traits of rice seeds of various cultivars. The rice seeds were photographed. Basic morphological traits, including width, length, aspect ratio, perimeter, and surface area, of the grains were then quantified using image process algorithms. The shape of the grains was characterized by elliptic Fourier descriptors. Principal component analysis was then applied to the descriptors for analyzing and visualizing the shape variation. The collected trait information and shape variation can be used for rice cultivar identification or for the study of phenotype-genotype association as future works. - Slides


Authors: Dr. Noppadol Kiriphet, National Electronics and Computer Technology Center, National Science and Technology Development Agency, Thailand



Effect of Planting Date and Locations on Characteristics Variation of Rice Varieties

Authors: Dr. Anchalee Prasertsak (remote speaker), Seed Technology Expert, Bureau of Rice Research and Development, Rice Department, Thailand - Slides

Agriculture with Satellite remote sensing & sensors

Authors: Dr Preesan Rakwatin (remote speaker), Geo-Informatics and Space Technology Development Agency (GISTDA), Thailand - Slides

Seating Arrangement : Classroom
Video Conferencing Facility : Yes
Remarks :