-
[show abstract]
[hide abstract]
ABSTRACT: Optical flow is a research topic of interest for many years. It has, until recently, been largely inapplicable to real-time applications due to its computationally expensive nature. This paper presents a new reliable flow technique which is combined with a motion detection algorithm, from stationary camera image streams, to allow flow-based analyses of moving entities, such as rigidity, in real-time. The combination of the optical flow analysis with motion detection technique greatly reduces the expensive computation of flow vectors as compared with standard approaches, rendering the method to be applicable in real-time implementation. This paper describes also the hardware implementation of a proposed pipelined system to estimate the flow vectors from image sequences in real time. This design can process 768 times 576 images at a very high frame rate that reaches to 156 fps in a single low cost FPGA chip, which is adequate for most real-time vision applications.
Radio Science Conference, 2009. NRSC 2009. National; 04/2009
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, we propose a robust and accurate algorithm based on a multimodal Sigma-Delta background estimation to extract the moving objects in image sequence of size 768 x 576 pixels taken from a static camera. Sigma-Delta estimation is used to compute two orders of temporal statistics for each pixel of the sequence providing a pixel-level decision framework. A serious limitation of this approach lies in the adaptation capability to certain complex scenes. In this paper, we avoid this limitation by modeling each pixel as mixture of three distributions to deal with complex scenes. We show that the enhanced performance is achieved by using the proposed algorithm. This paper describes also an FPGA-based implementation of the proposed algorithm at a very high frame rate that reaches to 1198 frames per second in a single low cost FPGA chip, which is adequate for most real-time vision applications.
Computer, Control and Communication, 2009. IC4 2009. 2nd International Conference on; 03/2009
-
[show abstract]
[hide abstract]
ABSTRACT: High-level computer vision tasks such as robot navigation, collision avoidance, path planning, and video surveillance require detection of the moving objects in the surrounding environment at real time. In the first part of this paper, a new algorithm based on a multi-modal distribution is presented to detect the moving objects in image sequence taken from a static camera with a small number of calculations. It is primarily composed of linear operations that are easily implemented in hardware and there is no iteration for any explicit coarse-to-fine control strategy. These properties make the real time flow of data possible through the hardware. We show that the elapsed time per frame is reduced by applying the proposed algorithm. For the second part, an FPGA-based implementation is described for the proposed algorithm at a very high frame rate that reaches to 1130 fps in a single low cost FPGA chip, which is adequate for most real-time vision applications. We show that the area of the implemented architecture can be reduced by 13.4%.
Mixed Design of Integrated Circuits and Systems, 2008. MIXDES 2008. 15th International Conference on; 07/2008
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, we propose a fast motion detection algorithm based on a multi-modal distribution to detect the moving objects in image sequence of size 768 × 576 pixels taken from a static camera with a small number of calculations to achieve a high frame rate. We show that the elapsed time per frame is reduced by applying our algorithm. Two major features make our algorithm a good candidate for hardware implementation. First, it is primarily composed of linear operations that are easily implemented in hardware. Linear operations can be represented efficiently in terms of the number of logic blocks required, and can be computed in one clock cycle. Second, there is no iteration for any explicit coarse-to-fine control strategy. This property makes the real time flow of data possible through the hardware. This paper describes also an FPGA-based implementation of the proposed motion detection algorithm at a very high frame rate that reaches to 1130 frames per second in a single FPGA chip, which is adequate for most real-time vision applications.
