Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Para optimizar los resultados de la búsqueda, se recomienda utilizar términos en inglés.

Análisis del contenido de imágenes

Resumen

Tipo:
Oferta Tecnológica
Referencia:
11 IT 53V2 3KC1
Publicado:
18/06/2015
Caducidad:
20/06/2015
Resumen:
Una empresa italiana ha desarrollado una tecnología de análisis del contenido de imágenes que consiste en procesar imágenes/vídeos para obtener explícitamente lo que se necesita saber de estas imágenes. La investigación en esta área permite el desarrollo de sistemas de recuperación de imágenes/vídeo basadas en contenidos, la detección y reconocimiento de objetos, la organización semántica de bases de datos de imágenes, el reconocimiento en sistemas de navegación robóticos, así como otras funcionalidades como marketing directo de datos multimedia, asistencia al conductor, etc. A diferencia de los sistemas convencionales, esta solución permite codificar y explotar el contenido visual de las imágenes para realizar diferentes tareas.

Details

Tittle:
Image processing technology for multimedia application
Summary:
An Italian research laboratory specialising in multimedia has developed a technology which consists of processing images/video of scenes by novel approaches based on holistic representations. This allows the development of systems for content-based image/video retrieval, object detection and recognition, semantic organization of image databases, place recognition for robot navigation systems, etc. Partners for license and/or technical cooperation agreements are sought.
Description:
An Italian research laboratory specialising in multimedia with specific competences in the field of image processing on consumer devices, non-photorealistic rendering, medical imaging and computer vision application, has developed a technology which consists of processing images/video of scenes by novel approaches based on holistic representations.

This technology allows representation models and machine learning techniques for:
- scene classification based on feature extracted on spatial or frequency domain;
- direct marketing learning and user profiling exploiting visual features;
- near duplicate image retrieval.

Current and potential domains of application are:
- automobile driver assistance
- digital photography
- industrial automation and inspection
- food analysis
- animal science
- robotics
- medical and biomedical
- security
- marketing
- forensic
- traffic and road management.

Partners are sought for license and/or technical cooperation agreements with the aim of implementing this technology in the market.
Current and Potential Domain of Application
- Automobile Driver Assistance
- Digital Photography
- Industrial automation and inspection
- Food Analysis
- Animal Science
- Robotics
- Medical and Biomedical
- Security
- Marketing
- Forensic
- Traffic and road management
Advantages and Innovations:
Differently from traditional systems (e.g., web search of images based on text as input), the proposed technology is able to encode and exploit the visual content of the images to deal with different tasks.

Humans are able to recognize complex visual scenes at a single glance, despite the number of objects with different poses, colors, shadows and textures. Recent studies suggest that the human visual system achieves its discrimination power using global information about the overall structure of the scene.

Many computer vision researchers have proved that holistic approaches can be efficiently used to solve the problem of rapid and automatic scene classification. In particular holistic approaches are able to recognize a scene bypassing the recognition of the objects and the details inside the scene.

Holistic representations of natural scenes is an effective and powerful source of information for semantic classification and analysis of arbitrary images. Novel approaches to classification of scenes by making use of holistic representations built on spatial and/or frequency domain have been experimented by the Italian research laboratory.
Stage of Development:
Prototype available for demonstration
IPs:
Secret know-how

Partner sought

Type and Role of Partner Sought:
- Type of partner sought: SMEs and companies.

- Specific area of activity of the partner: multimedia consumer devices and image processing.

- Task to be performed by the partner sought: implementation of the technology in the market.
Type of Partnership Considered:
Adaptation to specific needs
Joint further development
License Agreement
Technical consultancy
Testing of new applications
Transfer of knowledge in new raw materials

Client

Type and Size of Client:
R&D Institution
Already Engaged in Trans-National Cooperation:
No
Languages Spoken:
English

Keywords

Technology Keywords:
01003012 Imagen, procesado de imágenes, reconocimiento de modelos
08002001 Métodos de análisis y detección
01003007 Tecnología informática / gráficos, meta informática