Applied signal processing
Applied signal processing
The research is aimed at image and video processing, feature extraction, sensors and condition monitoring, with applications in, for example, forensics, acoustics, health technology, marine technology and remote control.
Applied signal processing is one of the most successful areas of research at BTH. Our theoretical and practical experiences that the group has brings new insights into international research and new solutions in the industry worldwide. Our knowledge in applied signal processing has resulted in approx. 25 patent families and hundreds of follow-up patents.
The research is usually conducted in close collaboration with industry. By bridging knowledge gained through research with our extensive experience of industrial cooperation new innovative solutions are created. We solve “real problems, in real life”, in collaboration with industry, institutions and the public sector.
The research in applied signal processing is mainly conducted at the Department of Mathematics and Natural Sciences.
Currently the research group conducts research within:
- Noise and vibration control
- Pattern recognition, Feature Extraction and Classification
- Condition Monitoring and Condition Based Maintenance
- Automatic Control
- Speech enhancement
- Acoustic measurement methods
- Sound and vibration analysis
- Distance laboratories with remote monitoring and control
- Image processing; Enhancement and measurement of image and video quality
Examples of projects
Enhancement and measurement of image and video quality
This project is part of a large field of research where the goal is to achieve higher perceptual quality of image and video for the human user of mobile equipment such as mobile phones. The work in the project is performed within some areas of interest as follows. In all of these areas consideration has to be made regarding low computing power and low memory usage.
VISIR aims at forming a group of cooperating universities and other organizations, a VISIR Consortium, creating software modules using open source technologies for online laboratories and/or setting up online lab stations. During the recent decades the amount of laboratory work in engineering education, etc. has been reduced. Basically the number of students has increased, while staff and funding resources have diminished. The VISIR project is an initiative to provide means for sustainable development in the education system as well as in the industry. By enabling access to experimental equipment for everyone everywhere as well as providing new tools, makes marketing and installation of new education facilities more effective. The VISIR project creates opportunities enabling the development of competitive actors on a global education market and it is based on an initiative to create open source software to be used for distributed online laboratories.
The field of Pattern Recognition (PR) explores ways to interpret and categorize patterns that emerge in our world by machine. In order to achieve this task there is a strong need for theoretical development as well as practical methods. Research in Pattern Recognition Lab (PRL) focus on development of new and enhanced algorithms to analyze patterns collected from sensors. A pattern recognition problem is typically solved using three fundamental actions after the information reaches the sensor: pre-processing, feature extraction and classification. The inputs to the PR systems explored here are typically extracted from a sensor to a digital form, for example a digital image.
The aim of the project is to improve acoustic communication solutions for audio conferencing units. An audio conference unit fundamentally consists of one or several loudspeakers and at least one microphone. The sound coming from the far end of the telephone line is transmitted into the room via the loudspeaker. The microphone is used to pick up the sound from the people in the room. The ideal behavior of the system is such that the sound emitted from the loudspeaker is identical to the original sound created in the far end, and that the microphone only picks up the sound from the people in the room.
While this ideal system is easily described the reality imposes several challenges. These challenges are caused by both physical and technological factors which directly or indirectly affects the ideal function. These factors can be divided into the following categories; acoustic echo, line echo, acoustic noise and room reverberation.
Noise and Vibration Control incl. Distance Laboratory
The research team at BTH is working with the development of a low cost hybrid silencer consisting of a module based system design and a digital feedfoward control system that fit the ventilation industries requirements. This module based low cost hybrid silencer is based on standard ventilation duct parts and passive silencers which is an advantage from the manufacture side. The module based design results in a very flexible silencer; the active part can be combined with different kinds of passive silencers and circular and square duct configurations. In these cases the passive silencer can be adapted to fit the specific control situation. The loudspeaker and microphones can also easily be reached outside the duct enabling easy maintenance of the system. The flexible design is unique for low cost hybrid silencers and such silencers have not been reported from other researchers. A research area in the active control of propagating sound in ducts is focusing on the control of higher order modes. To control the higher order modes a multiple channel active noise control system based on several loudspeakers and error sensors are needed. Current research in algorithm development focuses, for example, on tracking rapidly changing noise characteristics and reducing the calculation load for the control system with a view to be able to handle a large number of loudspeakers and error sensors for efficiently control the higher order modes.