Project-PECOViSiA – MORECoLAB – Montanhas de Investigação

Identification & Background

Project Name: PECOViSiA

Project Code: 18429_NORTE2030-FEDER-01477500 | Aviso MPr-2023-7

Main Objective: The Mirandesa cattle breed, originally from the Mirandese Plateau, had a relevant historical role in national livestock, being for decades essential in agricultural traction and valued for the high quality of its meat, highlighting the Mirandesa Posta. Currently, the breed is mainly intended for meat production, requiring strict control of growth parameters to ensure preservation and genetic improvement. However, traditional methods of identifying and weighing animals extensively are difficult to implement, cause high stress and have limitations in terms of reliability. This project proposes the development of a technological solution based on computer vision and artificial intelligence, integrated into a web platform and a mobile application, allowing the identification and estimation of the live weight of animals in a non-intrusive way. The solution aims to improve animal welfare, optimize daily management, support the genetic improvement of the Mirandesa breed and ensure consumers high quality meat.

Partners:
GEODOURO – CONSULTORIA E TOPOGRAFIA LDA (Leader)
COOPERATIVA AGRO-PECUÁRIA MIRANDESA CRL
INSTITUTO POLITÉCNICO DE BRAGANÇA
MORE – LABORATÓRIO COLABORATIVO MONTANHAS DE INVESTIGAÇÃO – ASSOCIAÇÃO
ASSOCIAÇÃO DOS CRIADORES DE BOVINOS DE RAÇA MIRANDESA

Start date: 02-01-2026
Completion date: 31-12-2028

Total eligible investment: 1 072 765,60 €
Funding supported: 873 881,20 €

The PECOViSiA project proposes to develop the aforementioned technologies covering the research and development phases in order to achieve 4 main objectives:

Develop an intelligent animal management system, with the integration of a Machine learning (ML) model and computer vision weighing at three different moments of the cattle’s life.
To develop a model composed of analysis methodologies that allow a morphological evaluation of Mirandesa cattle, with a view to improving the quality of the breed and the yield of its carcass
Create and train 1 artificial intelligence algorithm that allows the biometric recognition of Mirandesa cattle
Construction of a dataset of animal, body and snout images, 2D and 3D, whose objective at the end of the project is to make it publicly available to scientific and research entities

Activity 1: Research and Specifications.
Activity 2: Study and development of the concept.
Activity 3: Integration, Testing and Validation.
Activity 4: Promotion, Dissemination and Dissemination of Results.
Activity 5: Project Management and Technical Coordination.

Intelligent animal management system applied to the Autochthonous Mirandesa bovine breed
Biometric image identification system
Morphological evaluation model of cattle to improve breed quality and carcass yield
Artificial Intelligence algorithms for classification and prediction of cattle breed weights
dataset of animal, body and snout images, 2D and 3D

Identification & Background

MOrada

Contactos

horário de funcionamento

address

Contacts

working hours