No abstract

No abstract

Harvesting energy for IoT nodes in places that are permanently poorly lit is important, as many such places exist in buildings and other locations. The need for energy-autonomous devices working in such environments has so far received little attention. This work reports the design and test results of an energy-autonomous sensor node powered solely by solar cells. The system can cold-start and run in low light conditions (in this case 20 lux and below, using white LEDs as light sources). Four solar cells of 1 cm2 each are used, yielding a total active surface of 4 cm2. The system includes a capacitive sensor that acts as a touch detector, a crystal-accurate real-time clock (RTC), and a Cortex-M3-compatible microcontroller integrating a Bluetooth Low Energy radio (BLE) and the necessary stack for communication. A capacitor of 100 μF is used as energy storage. A low-power comparator monitors the level of the energy storage and powers up the system. The combination of the RTC and touch sensor enables the MCU load to be powered up periodically or using an asynchronous user touch activity. First tests have shown that the system can perform the basic work of cold-starting, sensing, and transmitting frames at +0 dBm, at illuminances as low as 5 lux. Harvesting starts earlier, meaning that the potential for full function below 5 lux is present. The system has also been tested with other light sources. The comparator is a test chip developed for energy harvesting. Other elements are off-the-shelf components. The use of commercially available devices, the reduced number of parts, and the absence of complex storage elements enable a small node to be built in the future, for use in constantly or intermittently poorly lit places.

No abstract

Image enhancement is mostly driven by intent and its future largely relies on our ability to map the space of intentions with the space of possible enhancements. Taking into account the semantic content of an image is an important step in this direction where contextual and aesthetic dimensions are also likely to have an important role. In this article we detail the state-of-the-art and some recent efforts in for semantic or content-dependent enhancement. Through a concrete example we also show how image understanding and image enhancement tools can be brought together. We show how the mapping between semantic space and enhancements can be learnt from user evaluations when the purpose is subjective quality measured by user preference. This is done by introducing a discretization of both spaces and notions of coherence, agreement and relevance to the user response. Another example illustrates the feasibility of solving the situation where the binary option of whether or not to enhance is considered.