From Poland Lab Wiki
Revision as of 01:57, 23 April 2016 by Trife (Talk | contribs)

Jump to: navigation, search

The PhenoApps project will converge novel advances in image processing and machine vision to deliver transformative mobile applications through established breeder networks. User-friendly mobile apps for field-based high-throughput phenotyping that utilize novel image analysis algorithms will be built and deployed to model and extract plant phenotypes. To ensure both immediate, broad deployment and functionality on a diverse set of crops, breeder networks for cassava and wheat will be engaged, providing a diverse set of target plant phenotypes, environments, breeding programs and working cultures.



Dramatic increases in the speed and ability to collect precision phenotypic data are needed to decipher plant genomes and accelerate plant breeding. Over the past decade, the availability of genomic data has exploded while the methods to collect phenotypes have made minimal advancements. This has led to a dramatic imbalance in data sets connecting genotype to phenotype and highlighting phenotyping as the remaining major bottleneck in plant breeding programs. This project will advance the field of 3D graphics and modeling, data mining, and deep learning through integration of simultaneous ground truth phenotypic measurements and imaging with mobile technology.

By focusing on novel algorithms delivered through mobile apps, innovative phenotyping tools can be rapidly deployed through readily available and highly-penetrant mobile technology. This approach will enable rapid dissemination and broad usability. Collectively equipping thousands of breeders around the world with tools for rapid collection, processing and analysis of complex phenotypes will provide the foundation for increasing genetic gain that will ultimately result in improved productivity, food security, nutrition, and income of smallholder farmers and their families in developing countries.


Add graphical app workflow (prelim data → algorithms → beta users → feedback and changes → implementation → beta test → release)

Project team

Add people and hierarchy chart (research teams, breeders, developers, etc.)



Wheat is one of the most important staple crops worldwide, providing 21% of the food calories and 20% of the protein for more than 4.5 billion people in 94 developing countries. In the developing world, wheat is a particularly important crop for food security as the primary staple for over 1.2 billion and an important food source for an additional 2.5 billion individuals living in poverty; men, women and children living on less than $2/day USD. In addition, wheat is the primary income source for some 30 million poor wheat farmers and their families.

In the developing world, the demand for wheat is projected to increase 60% by 2050. However, during this same period, increasing temperatures induced by climate-change are expected to reduce wheat production in developing countries by 20–30%. As a result of growing populations, demand for wheat is growing at an annual rate of 1.7% but wheat yields are growing at only about 1% annually. The increased demand is leading to market volatility and increased prices, which has a greater relative impact on the poor and food insecure. These combined factors are forecasted to double wheat prices. As a result, the purchasing power of poor consumers will decrease, leading to food insecurity and creating conditions for widespread social unrest. This scenario is aggravated by stagnating yields, soil degradation and loss of arable land, increasing fertilizer costs, loss of irrigation waters, and new virulent disease and pest strains.



Trait Priority Status
Spike count 0 None
Rust quantification 0 None
Seed size and shape 0 1KK
Leaf morphology 0 None
Plant architecture 0 None
Plant physiology 0 None



Cassava (Manihot esculenta Crantz.) is a highly adaptable starchy root crop and the primary staple food for more than 800 million people, largely in sub-Saharan Africa, which accounts for more than 50% of the total cassava production globally. This clonally propagated crop is also increasingly becoming a source of revenue from fresh and processed food, the production of starch-based products, biofuels and animal feed. The crop is cultivated in tropical and sub-tropical regions of Asia, Latin America and Africa. Currently, Nigeria is the largest world producer of cassava with a total output of 52.4 million MT, which was around 21% of the world total (252.2 million MT) in 2013. In Nigeria, cassava provides at least 80% of the daily caloric requirement for over 50% of the population. Similarly, cassava provides food and income to more than 80% of the small holder households in Uganda.

Its ability to grow in marginal environments with erratic rainfall, poor soil fertility and under low intensity management has made it one of the most important food security crops in the African continent. National and international cassava breeding efforts have made significant impact on cassava productivity, quality, and disease tolerance. However, cassava breeding continues to be a challenging, particularly as cassava is a highly heterogeneous and heterozygous vegetative propagated crop with variable flowering, low seed set, and a long breeding cycle. It takes several years from making a cross to conducting large-plot, replicated, field trials and selection of parents for the next cycle of germplasm improvement. As such, the rate of genetic gain per year for yield is very low (~1.2%).



Trait Priority Status
Root size and shape 0 1KK
Cassava mosaic disease 0 None
Cassava brown streak disease 0 None
Whitefly count 0 None
Leaf morphology 0 None
Plant architecture 0 None
Plant physiology 0 None


Field Book

Field Book was developed to eliminate paper note-taking from plant breeding programs and facilitate robust data collection and rapid data access. It is a standalone program that utilizes a straightforward interface that focuses on a single entry and trait at a time. The interface is dynamic and changes based on the type of trait being collected. Data can be analyzed the same day they are collected, resulting in the ability to find and fix any mistakes made when collecting data. Field Book has been adopted by many U.S. and international breeding programs. It is the primary data collection software used by the Triticeae Coordinated Agriculture Project, the NextGen Cassava project, many universities (KSU, Cornell, UNL, etc.), and even many private companies (Syngenta, Limagrain, and Bayer). As of April 2016, more than 1100 devices around the world have an active installation of Field Book.


1KK is an app designed to analyze seed lots. Its name comes from the one thousand kernel weight that is commonly used as a selection criteria in plant breeding programs. 1KK extracts seed morphology from images captured by phone and tablet cameras. A non-parametric algorithm is used to identify individual seeds for shape measurements. Reference circles of known size included on the background translate pixel measurements of seeds to actual size. Each individual seed length, width, and area are determined using the same algorithm implemented in SmartGrain. Data can be exported in a sample summary form and on a per-object basis. For measurement of thousand kernel weight, the total number of seeds are counted and divided by the total weight. For weight measurements, the app is compatible with Elane USB scales (1g resolution).


Inventory is an app designed to assist with rapid inventory and weighing of seed stocks. Inventory uses an Elane USB Scale to quickly weigh and categorize samples. In addition to Box and Sample ID, a timestamp and the name of the inventory person are also collected. Data is exported to a text file ready to be uploaded to a database.


Coordinate is a unified data collection app based on defining templates and then collecting data in grids created from those templates. Two templates are included by default: Seed Tray and DNA Plate. There are many customizations available when defining a new template including custom fields for grid metadata collection (e.g. Person, Date, etc.); the naming for rows and columns can be alphabetic or numeric; and rows, columns, or random cells can be excluded from data collection. All collected data is saved internally to the database and grids can be reloaded to continue collecting data or deleted if not needed.