Using fast Fourier transform (FFT) and fast wavelet transform (FWT) for thermal visualization in systems integrated with personal protective equipment

Project manager: Grzegorz Owczarek Ph.D. (Eng.)

Project summary:

The project aimed at developing numerical algorithms that would support thermal visualization in workplaces with decreased visibility.The preliminary stage of the project involved- analysing the way of obtaining thermal images in the context of information that could be used in analysing the environmental conditions and ways of integrating modules for thermal visualization with personal protective equipment,- determining the necessary requirements for processing information obtained with thermal visualization,- analysing the possibilities of using the algorithm of reverse Fourier transform with a high- and low-frequency function in thermal images,- preliminary algorithms and a computer program were developed for fast Fourier transform and fast wavelet transform.Then, sample fast Fourier transform and fast wavelet transform of the registered sequences of thermal images were conducted. Algorithms that made it possible to establish the number of people in the registered thermal image were also developed. On their basis, original software (ThermoSearch) was developed; it detects and counts people in the thermal image.The concept of a model of a system for thermal visualization was developed for places with decreased visibility. A model of this system was produced; it consists of three modules. The algorithm for recognizing people was also modified by introducing conditions for selecting areas of interest in the registered thermal image.

Unit: Department of Personal Protective Equipment

Duration: 01.01.2008 – 29.11.2010