ABSTRACT
Controlled measurements by a low-cost single-pixel microwave radiometer operating at 12.65 GHz were carried out to assess the detection and counting capability for targets warmer than the surroundings. The adopted reference test targets were pre-warmed water and oil; and a hand, both naked and wearing a glove. The results showed the reliability of microwave radiometry for counting operations under controlled conditions, and its effectiveness at detecting even warm targets masked by unheated dielectric layers.
An electromagnetic model describing the scenario sensed by the radiometer antenna is proposed, and comparison with the experimental observations shows a good agreement. The measurements prove that reliable counting is enabled by an antenna temperature increment, for each target sample added, of around 1 K.
Starting from this value, an analysis of the antenna filling factor was performed to provide an instrument useful for evaluating real applicability in many practical situations. This study also allows the direct people counting problem to be addressed, providing preliminary operational indications, reference numbers and experimental validation.
MW RADIOMETER OVERVIEW
Figure 1. Block diagram of the microwave radiometer operating at the center frequency of 12.65 GHz with a bandwidth of about 100 MHz. The receiver is based on a SAT-TV LNB
Experimental results were carried out by means of a portable low-cost MW radiometer operating at 12.65 GHz, developed and realized at the Department of Engineering of the University of Perugia, aiming at performing field measurements of fire spots in different environmental conditions. The radiometer scheme is shown in Figure 1: a detailed description of this sensor is provided, and only the main features will be summarized in the following.
Figure 3. Antenna temperature recorded during an experiment carried out to detect a fire spot
As shown in Figure 3, the relative maximum A was caused by a person inside the antenna beam when approaching to light the fire. Therefore, the radiometer detected the temperature increase due to the human body at a distance of 30 m from the antenna (a 60 cm offset dish antenna for SAT-TV application with β = 3◦, in this case).
SCENARIO
Figure 4. Scene sensed by a microwave radiometer; β is the antenna half-power beamwidth
This section deals with the model describing the microwave radiative contributions sensed by a radiometer observing a scenario of interest for counting applications. We consider a single-pixel MW radiometer with the antenna half-power beam β covering the relevant target area as in Figure 4.
RESULTS
Figure 5. Experimental test setup. The radiometer antenna is placed at a distance D from the target
Water and oil were put into a sample holder made of styrofoam, as shown in Figure 5, with a target-antenna placement determining a normal incidence. The styrofoam had the advantage of thermally insulating the samples during handling and measurements, and enables a microwave transmissivity better than 0.99, as tested.
Figure 17. Antenna temperature TA measured by the radiometer for the stationary situation depicted in the right inset
With the horn antenna having β = 30◦ and the distance antenna-wall D = 4 m, the antenna footprint area at the wall is around 3.7 m2. Three persons (2 adult females and 1 adult male) were positioned sequentially inside the footprint, standing stationary against the wall; the temporal evolution of the experiment is explained in Figure 17, in which the radiometer observations (TA) are shown in the following settings.
CONCLUSIONS
The experimental approach has proven the effectiveness of microwave radiometry for counting issues under controlled conditions, and the potential use of a low-cost/single-pixel radiometer as a promising sensor for gathering information related to the detection and counting of targets warmer than their surroundings. At the same time, the reliability of the electromagnetic model describing the scenario sensed by the radiometer antenna allowed for first guess estimation of the parameters required to design counting equipment.
For instance, the filling factor, which ensures a suitable antenna temperature increment for each sample, or the role of the temperature and emissivity of the target. Both the model and the preliminary experimental results (hand test) also demonstrated the possibility of tackling the people counting instance.
A great variety of actual situations, mainly related to the anthropometry of human beings and to the location of persons inside the antenna beam, may occur; in this work, we have analyzed some general cases and drawn typical operational indications, useful for future adjustments and tests. An experiment in an actual scenario reveals the MW radiometer as a promising technology in people counting issues, stimulating future investigations.
Source: University of Perugia
Authors: Federico Alimenti | Stefania Bonafoni | Luca Roselli
>> 200+ IoT Led Projects for Engineering Students