This paper presents an embedded board for rapid prototyping of real time imageprocessingalgorithms. The platform is based on the Texas Instruments' TMS320C6415, a new digital signal processor designed for high p...
详细信息
This paper presents an embedded board for rapid prototyping of real time imageprocessingalgorithms. The platform is based on the Texas Instruments' TMS320C6415, a new digital signal processor designed for high performance applications. Any analog video source, such as a camcorder or a vCR, can be used as an input signal. The images are captured, processed in real time and then displayed in a window of a graphical user interface. Both a mouse and a keyboard are available to interact with the system. A software environment is also provided thus allowing the rapid implementation of the algorithms in high-level language.
This paper describes the implementation of a custom DSP system to accelerate imageprocessingalgorithms used in the field of digital holography. The system, implemented on an FPGA platform, is intended for real-time ...
详细信息
This paper describes the implementation of a custom DSP system to accelerate imageprocessingalgorithms used in the field of digital holography. The system, implemented on an FPGA platform, is intended for real-time reconstruction of images captured on a large image sensor. Due to the large amount of processing information, it is not possible to perform a HDL simulation of a complete image reconstruction in reasonable time. Instead, a reconfigurable solution is being used for full scale image reconstruction, exhaustive testing of the functionality and for connecting the accelerator to external components, i.e. the image sensor, monitor output device and high-speed memory banks.
The paper proposes a new computationally efficient technique for DCT operation. Unlike related research, the technique reduces the number of computations by predicting the effect of quantization on DCT and avoiding ca...
详细信息
The paper proposes a new computationally efficient technique for DCT operation. Unlike related research, the technique reduces the number of computations by predicting the effect of quantization on DCT and avoiding calculations of those DCT values which lead to zero elements in the block after quantization. Experimental evaluation on a number of video benchmarks shows that our method is able to reduce the total number of computations by 29% for DCT and by 59% for quantization while maintaining high image quality.
In this paper we present a methodology for evaluating the operational effectiveness of facial recognition systems. After a brief introduction, we provide a description of the facial recognition process, its modes of o...
详细信息
In this paper we present a methodology for evaluating the operational effectiveness of facial recognition systems. After a brief introduction, we provide a description of the facial recognition process, its modes of operation (such as enrollment and operation), and signal processing operations (face capture, normalisation, feature extraction, database comparison and decision and action). We describe the methodology; from initial facial recognition system installation, tuning of the system, statistical analysis, to human factors and operational impact.
Synthetic aperture radar and ordinary radar systems must be optimized with respect to some quality criteria. A commonly used optimization gives minimum mean-square variance error. However, this optimization cannot tak...
详细信息
Synthetic aperture radar and ordinary radar systems must be optimized with respect to some quality criteria. A commonly used optimization gives minimum mean-square variance error. However, this optimization cannot take into account all characteristics of the radar image, such as speckle. This leads to a number of non-optimal secondary processingalgorithms (nonlinear filtering, statistical estimation, etc.). The paper presents common quality criteria applicable to space and space-time processing. Radar image formation errors are separated into three kinds: dynamic, fluctuation and noise errors. Dynamic errors are caused by smoothing the radar space ambiguity function. One realisation of the stochastic process observed during the finite time interval leads to inconsistent estimation, which is called fluctuation error (speckle noise). Noise error is caused by additional noise present in the input. The paper proposes statistical criteria applicable to optimizing the processingalgorithms in radar systems.
We present an approach for large medical images database indexing and retrieval using TTA10 algorithm. Our approach consist of primary and secondary tessellation stages, therefore it can be used in computer cluster wi...
详细信息
We present an approach for large medical images database indexing and retrieval using TTA10 algorithm. Our approach consist of primary and secondary tessellation stages, therefore it can be used in computer cluster with parallel multiprocessor architecture. We propose compact index structure, which increases systems scalability and achieves logarithmic complexity. Tree structured index in tree tessellation algorithm v.10 (TTA10) is optimized for balance monitoring and real-time subtree reorganization, avoiding of changes in the structure of the entire tree.
The application of different interpolation algorithms after image decimation, as a precompression or compression stage, is presented. Interframe and intraframe horizontal, vertical, cross and adaptive interpolation, w...
详细信息
The application of different interpolation algorithms after image decimation, as a precompression or compression stage, is presented. Interframe and intraframe horizontal, vertical, cross and adaptive interpolation, with and without additional information transmission, are observed in order to minimise the realisation complexity and to achieve the best quality of image restoration.
Biomedical Engineering is a relatively new interdisciplinary science. This paper presents the biomedical engineering activity, which is carried out at Budapest University of Technology and Economics and its partner in...
详细信息
Biomedical Engineering is a relatively new interdisciplinary science. This paper presents the biomedical engineering activity, which is carried out at Budapest University of Technology and Economics and its partner institutes. In the first part the main goals and the curriculum of the Biomedical Engineering Education Program (BMEEP) is presented. The second part of the paper summarizes the most important biomedical engineering researches carried out mostly in the Biomedical Engineering Laboratory of our university.
The author has applied a guided scrambling (GS) technique, usually used for a recording channel, to digital watermarking (Kunisa, A., Proc. SPIE: Security and Watermarking of Multimedia Contents, v, 2003). The paper m...
详细信息
The author has applied a guided scrambling (GS) technique, usually used for a recording channel, to digital watermarking (Kunisa, A., Proc. SPIE: Security and Watermarking of Multimedia Contents, v, 2003). The paper modifies the GS-based watermarking algorithm to a block-wise watermarking system. It is well suited for compression processing, such as JPEG or JPEG2000, and is readily applied to digital cameras.
In this paper, a multiplexed multiplier architecture (MMA) for a field programmable gate array (FPGA) implementation of the least mean square (LMS) adaptive filter is developed and presented. In the proposed architect...
详细信息
In this paper, a multiplexed multiplier architecture (MMA) for a field programmable gate array (FPGA) implementation of the least mean square (LMS) adaptive filter is developed and presented. In the proposed architecture, hardware multipliers are reused, i.e. multiplexed in time, for both filtering and adaptation. The number of multipliers may be chosen to achieve certain design trade-offs. The design trade-offs considered in this paper include on-chip area, filter size, maximum filter throughput, and power consumption.
暂无评论