The goal of smart cities is to improve efficiencies, enhance sustainability, advance quality of life and reduce energy consumption. One of the key factors to accomplish a smart city involves the use of IoT and information technology infrastructure which can be described as the foundation of a smart city. Its effective implementation will allow the city to meet its wide range of requirements while being able to respond to innovations, such as advanced sensors, analytic tools, measurement, and artificial intelligent based solutions. This paper investigates and compares between the use of low and high resolution infrared sensors as part of the Internet of Things (IoT) to estimate crowds in cities to enhance and optimise the efficiency of the transportation process and other public services for low density scenarios. A case study was conducted in Nottingham city at one of the tram stops. An experimental methodology is used where different number of people are captured and the results are compared using different image processing techniques. The findings show that both technologies are useful in the estimation of crowd density, however, the high resolution camera has been found to be more accurate in estimating the number of people albeit it is more expensive for the integration into infrastructures. The practical implication is that low-cost and low resolution infrared cameras could provide reasonable results. However, for higher accuracy, high resolution infrared cameras will be needed; and they are potentially more expensive. So a compromise might be needed between cost and performance to encourage the installation of more IoT systems using infrared technologies.