Wireless nodes with high-resolution cameras are enabling wide range of new applications for wireless sensor networks, while a special design approach is required in order to save energy and prolong lifetime of a battery-powered wireless sensor nodes. To be able to assess an energy efficiency of data-intensive wireless transmission, a measurement-based model of the wireless sensor node is introduced. The node model implementation represents a typical architecture of the sensor node, where finite state machines are used to describe behaviour of each unit, making it possible to tune behaviour of the node model to different scenarios. A discrete event simulator OMNET++ was used for the node model implementation. In simulations, the node model was set to reflect behaviour of commercially available ZigBee (CC2530) and Wi-Fi (CC3000) modules. Our results show that in scenarios when the node sends up to 64~kB of data, the ZigBee node provides longer lifetime, while for higher data amount Wi-Fi provide longer lifetime. As energy consumption of the wireless transfer is proportional to the amount of data the node has to sent, the primary aim of an employed image processing is to reduce the amount of data. Analysis of suitable change detection algorithms, that includes performance and energy efficiency analysis, was conducted on several image sequences, that were acquired in different illumination conditions. Results indicate that a change of illumination conditions can have significant effect on algorithms performance and that usage of an artificial scene illumination provides results with higher change detection efficacy. Energy consumption measurements of algorithms execution have revealed that more complex algorithms, providing higher change detection performance, but taking more operations to execute, requires significantly more energy than simple algorithms. In all cases the change detection algorithms were able to significantly reduce amount of data the nodes have to send. A comparison of an overall energy consumption of the node sending the high resolution image and the energy consumption required for the image processing reveals that the image processing is required to achieve long lifetime. The lowest energy consumption is achieved when the artificial scene illumination and simple change detection algorithms are employed. In case of ZigBee node, the more complex image processing, that provides the lowest data amount the node has to sent, results with lower overall energy consumption, while in case of Wi-Fi node overall lower energy consumption is achieved if image dilatation and connected components labelling are not executed. For a proposed node architecture, the lifetime of ZigBee node and Wi-Fi node differ only for approximately 5%.