Image feature detection and matching is a fundamental operation in image processing. As the detected and matched features are used as input data for high-level computer vision algorithms, the matching accuracy directly influences the quality of the results of the whole computer vision system. Moreover, as the algorithms are frequently used as a part of a real-time processing pipeline, the speed at which the input image data are handled is also a concern. The paper proposes an embedded system architecture for feature detection and matching. The architecture implements the FAST feature detector and the BRIEF feature descriptor and is capable of establishing key point correspondences in the input image data stream coming from either an external sensor or memory at a speed of hundreds of frames per second, so that it can cope with most demanding applications. Moreover, the proposed design is highly flexible and configurable, and facilitates the trade-off between the processing speed and programmable logic resource utilization. All the designed hardware blocks are designed to use standard, widely adopted hardware interfaces based on the AMBA AXI4 interface protocol and are connected using an underlying direct memory access (DMA) architecture, enabling bottleneck-free inter-component data transfers.

Using passive infrared sensors is a well-established technique of presence monitoring. While it can significantly reduce energy consumption, more savings can be made when utilising more modern sensor solutions coupled with machine learning algorithms. This paper proposes an improved method of presence monitoring, which can accurately derive the number of people in the area supervised with a low-cost and low-energy thermal imaging sensor. The method utilises U-Net-like convolutional neural network architecture and has a low parameter count, and therefore can be used in embedded scenarios. Instead of providing simple, binary information, it learns to estimate the occupancy density function with the person count and approximate location, allowing the system to become considerably more flexible. The tests show that the method compares favourably to the state of the art solutions, achieving significantly better results.

This paper presents the complete calibration procedure of a multi-camera system for mobile robot motion registration. Optimization-based, purely visual methods for the estimation of the relative poses of the motion registration system cameras, as well as the relative poses of the cameras and markers placed on the mobile robot were proposed. The introduced methods were applied to the calibration of the system and the quality of the obtained results was evaluated. The obtained results compare favourably with the state of the art solutions, allowing the use of the considered motion registration system for the accurate reconstruction of the mobile robot trajectory and to register new datasets suitable for the benchmarking of indoor, visual-based navigation algorithms.

Although the introduction of commercial RGB-D sensors has enabled significant progress in the visual navigation methods for mobile robots, the structured-light-based sensors, like Microsoft Kinect and Asus Xtion Pro Live, have some important limitations with respect to their range, field of view, and depth measurements accuracy. The recent introduction of the second- generation Kinect, which is based on the time-of-flight measurement principle, brought to the robotics and computer vision researchers a sensor that overcomes some of these limitations. However, as the new Kinect is, just like the older one, intended for computer games and human motion capture rather than for navigation, it is unclear how much the navigation methods, such as visual odometry and SLAM, can benefit from the improved parameters. While there are many publicly available RGB-D data sets, only few of them provide ground truth information necessary for evaluating navigation methods, and to the best of our knowledge, none of them contains sequences registered with the new version of Kinect. Therefore, this paper describes a new RGB-D data set, which is a first attempt to systematically evaluate the indoor navigation algorithms on data from two different sensors in the same environment and along the same trajectories. This data set contains synchronized RGB-D frames from both sensors and the appropriate ground truth from an external motion capture system based on distributed cameras. We describe in details the data registration procedure and then evaluate our RGB-D visual odometry algorithm on the obtained sequences, investigating how the specific properties and limitations of both sensors influence the performance of this navigation method.