Dilepix Blog

[PRESS] AI for a more responsible production of animal proteins

Written by Aurelie Busnel | 8 July 2021

PRESS RELEASE 


Rennes, June 23, 2021. In full growth, Dilepix, the Rennes-based startup, expert in artificial intelligence & computer vision, reinforces its commitment to a production of animal proteins that is more respectful of animal welfare while having a low environmental footprint.

Dilepix is an impact startup that provides farmers and agricultural professionals (veterinarians, technicians, consultants) with solutions to drastically reduce the use of chemicals by helping them better monitor the health of their animals and avoid systematic preventive treatments (antibiotics and hormones).


Two reasons for this:

  1. Population growth: How to combine the increase in the world's population with the decrease in agricultural land?

    According to UN projections, the world's population is expected to increase by 2 billion over the next 30 years (from 7.7 billion today to 9.7 billion in 2050) while agricultural land is shrinking due to urban sprawl and soil artificialisation. Countries with rapidly growing populations (Asia, China, sub-Saharan Africa) are so hungry for protein that global demand is expected to grow by 40% by 2030 (with a 33% increase for animal protein).


  2. Societal pressure for animal welfare

    Consumers and actors in the agricultural sector, increasingly sensitive to animal welfare, want to ensure good breeding conditions while reducing their environmental footprint.


How does Dilepix intend to respond to these two challenges?

On the one hand, to encourage the pig industry to improve animal welfare

 

Dilepix helps the current animal protein production sectors, in particular pig production, to move towards more virtuous breeding conditions. Awareness of animal welfare is pushing the entire industry to change its practices towards measuring health status at the individual level and no longer at the batch level.

Indeed, due to the difficulties in identifying the animals, treatments are still currently carried out on a collective scale.

Dilepix software solutions target the right animal at the right time and measure situations that were previously impossible! For example, AI and computer vision make it possible to monitor the activity of each pig and to individually adjust their feed according to their energy and physiological needs.

"With a solution like ours, breeders and veterinarians are able to measure the health level of each animal individually," says Alban Pobla (co-founder). "And this is precisely what seduced the research institutes (INRAE - IFIP), the pharmaceutical group Boehringer Ingelheim (within the framework of the Synapse program), as well as a veterinary group to work with us to develop modules aimed at improving the health of pigs." 

 

Four modules for monitoring and automatic detection of pigs

 

  1. Automatic detection of piglet tail biting to comply with the docking ban.

  2. The estimation of the weight of pigs in post-weaning and fattening

  3. Optimization of sow reproduction to drastically reduce the use of hormones (NB: This innovation will be presented at Space 2021)

  4. The automatic measurement of the pigs' activity which tends towards an individualized ration system.

 


And secondly, to accelerate the production of alternative proteins by automating the production lines of insect factories.

 

 

The FAO estimates that world food production will have to increase by 70% and meat production (pork, beef, poultry) will have to double to feed the world's inhabitants. This demographic challenge raises a number of questions, in particular that of the increase in animal protein requirements.

Still emerging in our Western countries, proteins from insects are a promising alternative, especially in aquaculture, pet food and poultry farming.

This very young sector represents a gigantic market (the global weight should reach 1.10 billion euros by 2025, against 110 million euros today) whose capacities will be insufficient for the next 10 years.

But in order to produce sufficient quantities, while respecting quality standards, insect producers will have to optimize their production process by moving from the manual stage to the industrial stage.
This transition to an industrial scale will only be possible by automating the production processes.

Because Dilepix is a startup concerned about its impact on the environment and believes in this alternative production method, it decided to put its technology at the service of entomoculture.

Today, it is the only startup to help insect producers to produce better. The software solution it has developed automates quality analyses on the production line and allows the creation of precise indicators.

 

Four modules have been developed to accelerate the control of insect production

  1. Automatic counting of larvae in a few seconds even when they are clustered

  2. Estimation of larval mortality rate

  3. Instant evaluation of production batches (growth stages, purity...)

  4. Measuring the activity of the larvae

About Dilepix
Dilepix, a startup from Rennes, France, expert in artificial intelligence and computer vision, develops software solutions to automatically monitor the activity of pigs and insects. To achieve this, it relies on the images/videos produced by cameras placed in the livestock buildings or embedded in the production lines. The algorithms (deep learning) detect situations of interest that require special monitoring or specific action.


Founded by Aurélien Yol, a former INRIA engineer and Alban Pobla, a serial entrepreneur, Dilepix was born in 2018. Hosted and supported by INRIA, then by the Village by CA Ille-et-Vilaine until 2020, the start-up is now flying under its own steam and has 13 employees.

 

From left to right : Aurélien YOL (CTO) & Alban POBLA (CEO)

 

Dilepix's know-how combines an expertise in artificial intelligence (neural networks developed internally by its engineers) with advanced skills in image analysis. The combination of these two technological bricks allows an ultra-precise detection and localization of agronomic threats and opportunities. Developed in laboratory and tested by major industrialists, the technology used by Dilepix is patented.

Dilepix was a winner of the prestigious iLab competition in 2018 and participated two years in a row (2019 and 2020) at CES in Las Vegas.

It completed an initial seed round of €1.7 million in 2018 and is about to launch a new Series A round.

Dilepix is derived from the scientific name of the chameleon "dilepis", an animal with excellent visual skills.

Contact Presse : aurelie.busnel@dilepix.com