MENU

R&d for integrated artificial intelligent system for detecting the wildlife migration

Integrated artificial intelligent system for detecting the wildlife migration that will utilise video surveillance technology with computer vision technology, that will provide gov. Structural and eco. Organisations with accurate data about wildlife migration for optimising road network.

Tasks of image recognition are currently being addressed in many fields of human activity. At the same time both the government and NGOs exerted considerable pressure on the environment protection. They spent considerable resources on environmental protection of various animal species, that are endangered in their natural environment, particularly by the building of road structures. In connection with this issue and the need to optimize the spending of these funds, there is a need to ensure monitoring of animals migration, especially around the intersection of their migration routes with highways, already finished or under construction, which could potentially limit migration, or completely block. In recent years were in the CR built special crossings for wildlife costing about 3 000 000 000 Kč. But, as it now appears, most these so-called ecoducts are not adequately functioning. This failure even admits Road and Motorway Directorate (RMD) - an institution which built them. Instead of this expensive ecoducts in the future they intends to build a so-called migration corridors. Planting a forest instead of building bridges is much cheaper. Road managers want by their new tactic to save 15 billion Czech Crowns till 2025. Migration corridors should be build for newly constructed roads and for the older ones also. But how to know where to animals migrate? And this will solve our project. It is planned to monitor the pattern of migration for at least 12 years, to verify its proper functionality. We will develop an integrated system with elements of artificial intelligence to monitor the movement of animals which will provide data of wildlife migration in designated area. This data are currently provided by standard methods (direct observation, field tracks, droppings, calling observations, and others) that can not cover a continuous period of time and even then, this is very time consuming. Alternatively, for collection of these data are used CCTV systems running continuously. However , due to the large amount of ballast records, evaluation is too time consuming. Our project will consist of: Analysis of the suitable hardware and software resources for the project. Ethology research of different species of wild animals. Research and development of cognitive strategies for the identification of species of wild animals. Research and development of an integrated system with elements of artificial intelligence to monitor the movements of wild animals. Research and development of database system for processing the outputs. The verification of developed system. Application for the verification of practical results, determining the effect of the migration potential of migratory objects.
Acronym: 
DETECTGAME
Project ID: 
6 752
Start date: 
01-03-2012
Project Duration: 
46months
Project costs: 
1 150 000.00€
Technological Area: 
Imaging, Image Processing, Pattern Recognition
Market Area: 
Animal Husbandry

Raising the productivity and competitiveness of European businesses through technology. Boosting national economies on the international market, and strengthening the basis for sustainable prosperity and employment.