Håkan Ardö

Bio: Håkan Ardö is an academic researcher from Lund University. The author has contributed to research in topics: Hidden Markov model & Segmentation. The author has an hindex of 12, co-authored 47 publications receiving 566 citations.

A Hardware Architecture for Real-Time Video Segmentation Utilizing Memory Reduction Techniques

Abstract: This paper presents the implementation of a video segmentation unit used for embedded automated video surveillance systems. Various aspects of the underlying segmentation algorithm are explored and modifications are made with potential improvements of segmentation results and hardware efficiency. In addition, to achieve real-time performance with high resolution video streams, a dedicated hardware architecture with streamlined dataflow and memory access reduction schemes are developed. The whole system is implemented on a Xilinx field-programmable gate array platform, capable of real-time segmentation with VGA resolution at 25 frames per second. Substantial memory bandwidth reduction of more than 70% is achieved by utilizing pixel locality as well as wordlength reduction. The hardware platform is intended as a real-time testbench, especially for observations of long term effects with different parameter settings.

Application of automated video analysis for behavioural studies: concept and experience

Abstract: Lund University, Sweden, is developing a video analysis system for making long-term behavioural studies, primarily in complex urban environments. Road users are detected using the KLT (Kanade-Lucas-Tomasi) interest point tracker. Trajectories are estimated using foreground–background segmentation, whereas speed is estimated using the shape analysis of interest points. The extracted trajectories are further used for behavioural analysis. The authors present the experience from an ongoing study in Stockholm city, where the task was to find out if allowing two-way bicycle traffic on one-way streets had negative effects on safety. The video analysis system was applied to detect biking in the ‘wrong’ direction and analyse traffic conflicts between cyclists and other road users. The manual observations done in parallel allowed validating the accuracy of system performance.

Development of automatic surveillance of animal behaviour and welfare using image analysis and machine learned segmentation technique.

Abstract: In this paper the feasibility to extract the proportion of pigs located in different areas of a pig pen by advanced image analysis technique is explored and discussed for possible applications. For example, pigs generally locate themselves in the wet dunging area at high ambient temperatures in order to avoid heat stress, as wetting the body surface is the major path to dissipate the heat by evaporation. Thus, the portion of pigs in the dunging area and resting area, respectively, could be used as an indicator of failure of controlling the climate in the pig environment as pigs are not supposed to rest in the dunging area. The computer vision methodology utilizes a learning based segmentation approach using several features extracted from the image. The learning based approach applied is based on extended state-of-the-art features in combination with a structured prediction framework based on a logistic regression solver using elastic net regularization. In addition, the method is able to produce a probability per pixel rather than form a hard decision. This overcomes some of the limitations found in a setup using grey-scale information only. The pig pen is a difficult imaging environment because of challenging lighting conditions like shadows, poor lighting and poor contrast between pig and background. In order to test practical conditions, a pen containing nine young pigs was filmed from a top view perspective by an Axis M3006 camera with a resolution of 640×480 in three, 10-min sessions under different lighting conditions. The results indicate that a learning based method improves, in comparison with greyscale methods, the possibility to reliable identify proportions of pigs in different areas of the pen. Pigs with a changed behaviour (location) in the pen may indicate changed climate conditions. Changed individual behaviour may also indicate inferior health or acute illness.

Towards safe human-robot interaction in robotic cells: An approach based on visual tracking and intention estimation

Abstract: Removing the safety fences that separate humans and robots, to allow for an effective human-robot interaction, requires innovative safety control systems. An advanced functionality of a safety controller might be to detect the presence of humans entering the robotic cell and to estimate their intention, in order to enforce an effective safety reaction. This paper proposes advanced algorithms for cognitive vision, empowered by a dynamic model of human walking, for detection and tracking of humans. Intention estimation is then addressed as the problem of predicting online the trajectory of the human, given a set of trajectories of walking people learnt offline using an unsupervised classification algorithm. Results of the application of the presented approach to a large number of experiments on volunteers are also reported.

Hardware accelerator design for video segmentation with multi-modal background modelling

Abstract: Among many of the algorithms for video segmentation, one based on a statistical background model (Stauffer, C. and Grimson, W., Proc. IEEE Conf. Computer Vision and Pattern Recognition, 1999) was developed with the unique feature of robustness in multi-modal background scenarios. However, with a large number of calculations due to the pixel-wise processing of each frame, such an algorithm could only achieve a low frame rate, far from real-time requirements, on computers. A hardware accelerator is proposed, with a dedicated architecture aimed at addressing both computation and memory bandwidth demands. The whole system is targeted to an FPGA platform, which serves as a real-time test bench where long term effects caused by fixed point quantization and various parameter settings can be studied. Meanwhile, memory bandwidth as well as memory size are investigated, and reduction by up to 60 percent, through similarity exploitation for neighboring Gaussian parameters, is envisioned. Furthermore, a controller synthesis tool is used to relieve the effort for the manual design of the complex control unit which schedules the operations of the whole system.

Background Modeling using Mixture of Gaussians for Foreground Detection - A Survey

Abstract: Mixture of Gaussians is a widely used approach for background modeling to detect moving objects from static cameras. Numerous improvements of the original method developed by Stauffer and Grimson [1] have been proposed over the recent years and the purpose of this paper is to provide a survey and an original classification of these improvements. We also discuss relevant issues to reduce the computation time. Firstly, the original MOG are reminded and discussed following the challenges met in video sequences. Then, we categorize the different improvements found in the literature. We have classified them in term of strategies used to improve the original MOG and we have discussed them in term of the critical situations they claim to handle. After analyzing the strategies and identifying their limitations, we conclude with several promising directions for future research.

Progress and prospects of the human---robot collaboration

Abstract: Recent technological advances in hardware design of the robotic platforms enabled the implementation of various control modalities for improved interactions with humans and unstructured environments. An important application area for the integration of robots with such advanced interaction capabilities is human---robot collaboration. This aspect represents high socio-economic impacts and maintains the sense of purpose of the involved people, as the robots do not completely replace the humans from the work process. The research community's recent surge of interest in this area has been devoted to the implementation of various methodologies to achieve intuitive and seamless human---robot-environment interactions by incorporating the collaborative partners' superior capabilities, e.g. human's cognitive and robot's physical power generation capacity. In fact, the main purpose of this paper is to review the state-of-the-art on intermediate human---robot interfaces (bi-directional), robot control modalities, system stability, benchmarking and relevant use cases, and to extend views on the required future developments in the realm of human---robot collaboration.

Evaluation of traffic safety, based on micro-level behavioural data: Theoretical framework and first implementation

Abstract: A traffic encounter between individual road users is a process of continuous interplay over time and space and may be seen as an elementary event with the potential to develop into an accident. This paper proposes a framework for organising all traffic encounters into a severity hierarchy based on some operational severity measure. A severity hierarchy provides a description of the safety situation and trade-off between safety and efficiency in the traffic system. As a first approach to study the encounter process, a set of indicators is proposed to describe an encounter. These indicators allow for a continuous description even if the relationship between the road users changes during the process (e.g., when they are on a collision course or leave it). Automated video analysis is suggested as a tool that will allow data collection for validation of the proposed theories.

Recent Advanced Statistical Background Modeling for Foreground Detection - A Systematic Survey

Abstract: Background modeling is currently used to detect moving objects in video acquired from static cameras. Numerous statistical methods have been developed over the recent years. The aim of this paper is firstly to provide an extended and updated survey of the recent researches and patents which concern statistical background modeling and secondly to achieve a comparative evaluation. For this, we firstly classified the statistical methods in terms of category. Then, the original methods are reminded and discussed following the challenges met in video sequences. We classified their respective improvements in terms of strategies used. Furthermore, we discussed them in terms of the critical situations they claim to handle. Finally, we conclude with several promising directions for future research. The survey also discussed relevant patents.