Sorry, you need to enable JavaScript to visit this website.

Back to Observatory

Larvia: A shrimp larvae analysis app using Deep Learning

Description of the service

Larvia (Larva + AI, in Spanish), is a counting, individual weight estimation, weight grading, uniformity estimation, health status estimation, and geolocation of aquaculture samples service, including larvae and juvenile in raceways. In addition, in our web dashboard, our users can follow up on their seeding/reviews, and through business intelligence tools, make better business and production decisions.

Problem that it solves

Technicians usually count larvae or juveniles individually only to obtain the number of animals per gram. This process takes 3 to 5 minutes and is subject to human error. In less than 15 seconds, Larvia allows users to retrieve larvae size data, plus a histogram of weights, lengths, average weight, average length, coefficient of variation, uniformity, number of animals per size, pigmentation histogram, and georeferenced location, as well as images to validate detection accuracy. This service also gives access to a web dashboard with business intelligence tools for decision-making.

Type of AI app used

Events detection (anomaly detection and early warning), Recognition

Main results to June 30, 2021

The quality of detection has been improved, from having only larvae to having larvae and juveniles.

Three main bottlenecks faced during implementation

Data labeling

Lessons learned in the design or use of AI for social impact

These tools streamline production by enabling better decision-making.

Country of origin

Ecuador

Geographic scope of operations

Ecuador, Peru, Mexico, Brazil, US Nicaragua, Honduras, Saudi Arabia, Thailand

Type of executing entity

Startup

Sector/industry

Aquaculture
Agriculture

Sustainable Development Goal(s) to which your AI solution contributes

2 (zero hunger)

8 (decent work and economic growth)

9 (industry, innovation, and infrastructure)

IA app developed internally or by a third party

Internally

Name of implementing entity

Larvia SA

Stakeholders involved

Private entrepreneurs: Ivan Ramirez, Chris Olsen, Jaime Rodriguez

Percentage of the development team that are women

20%

Year they started using AI-based models

2018
It may interest you
Tools

Cada vez más, actores públicos y privados se plantean cómo escalar su impacto a través del uso de la tecnología.

Publications

The Regional Landscape and 12 Country Snapshots

Publications

Junto con el MIT desarrollamos un estudio en el que utilizamos por primera vez herramientas de inteligencia artificial para elegir a beneficiarios de programas sociales