Alexander Pichler

Automatic pedestrian detection and counting applied to urban planning

The statistical knowledge of human flows in the streets is mandatory for urban planning. Today many cities use the expensive method of manual pedestrian counting, since there is no reliable automatic counting device. This project aims at achieving the first efficient, real-time, embedded and autonomous system that provides high-level data. Our first work focused on the development of a reliable counting method under Matlab™. On the basis of video sequences recorded in the city of Mulhouse we have validated the robustness of our approach.

Bio-inspired approach for intelligent unattended ground sensors

Improving the surveillance capacity over wide zones requires a set of smart battery-powered Unattended Ground Sensors capable of issuing an alarm to a decision-making center. Only high-level information has to be sent when a relevant suspicious situation occurs. In this paper we propose an innovative bio-inspired approach that mimics the human bi-modal vision mechanism and the parallel processing ability of the human brain. The designed prototype exploits two levels of analysis: a low-level panoramic motion analysis, the peripheral vision, and a high-level event-focused analysis, the foveal vision. By tracking moving objects and fusing multiple criteria (size, speed, trajectory, etc.), the peripheral vision module acts as a fast relevant event detector. The foveal vision module focuses on the detected events to extract more detailed features (texture, color, shape, etc.) in order to improve the recognition efficiency. The implemented recognition core is able to acquire human knowledge and to classify in real-time a huge amount of heterogeneous data thanks to its natively parallel hardware structure. This UGS prototype validates our system approach under laboratory tests. The peripheral analysis module demonstrates a low false alarm rate whereas the foveal vision correctly focuses on the detected events. A parallel FPGA implementation of the recognition core succeeds in fulfilling the embedded application requirements. These results are paving the way of future reconfigurable virtual field agents. By locally processing the data and sending only high-level information, their energy requirements and electromagnetic signature are optimized. Moreover, the embedded Artificial Intelligence core enables these bio-inspired systems to recognize and learn new significant events. By duplicating human expertise in potentially hazardous places, our miniature visual event detector will allow early warning and contribute to better human decision making.

Novel All-Hardware Closed-Loop Adaptive Optics System using Massively-Parallel Neural Processing

In most applications of laser technology and optics the beam quality, the ability to focus a laser beam and the achievement of a good optical resolution play an important role. For the compensation of distortions adaptive optics is used. Classical adaptive-optics control schemes use matrix operations, which show a non-linear computation time dependence with matrix size, making it difficult to achieve high control loop frequencies at high resolution. A novel closed-loop adaptive optics is presented using a massively-parallel neural network in an all-hardware setup. It can be used for a fast real-time wave front sensor and for closed-loop operation.