A Custom Image Convolution DSP with a Sustained Calculation Capacity of >1 GMAC/s and Low I/O Bandwidth

A customized processor for real time image convolution has been designed to increase the performance of an instrument for automated cereal grain quality assessment. Image convolution requires an extensive amount of calculation capacity and a corresponding amount of data transfers, hard to achieve with standard processors in real time. Therefore, a tailored architecture with a streamlined dataflow has been developed with emphasis on a system design perspective. The designed processor has a sustained calculation capacity of >1 GMAC/s and on-chip line buffers reduce the amount of external data transfers. Hence, the complexity of the designed processor has been increased to gain a lower complexity of the complete system. To achieve powerful and versatile filtering the size of the programmable kernel functions have been maximized to 15 × 15.

[1]  Anders Krogh,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[2]  F. Lai,et al.  Discrimination of wheat and nonwheat components in grain samples by image analysis , 1989 .

[3]  R. W. Brodersen,et al.  Architectures and design techniques for real-time image-processing IC's , 1987 .

[4]  Lars E. Thon,et al.  From C to Silicon , 1992 .

[5]  Jan M. Rabaey,et al.  An integrated CAD system for algorithm-specific IC design , 1991, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[6]  E. Shwedyk,et al.  Discrimination of wheat class and variety by digital image analysis of whole grain samples , 1987 .

[7]  M. Faulkner,et al.  Adaptive linearisation using pre-distortion , 1990, 40th IEEE Conference on Vehicular Technology.

[8]  Jae S. Lim,et al.  Two-Dimensional Signal and Image Processing , 1989 .

[9]  Viktor Öwall,et al.  An algorithm specific DSP for automatic control of a high speed power converter , 1995 .

[10]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[11]  Viktor Öwall,et al.  A custom DSP for feature extraction with a calculation capacity of >1 GMAC/s and low I/O bandwidth , 1997 .

[12]  M. Torkelson,et al.  Custom DSP design of a GSM speech coder , 1995 .

[13]  Viktor Öwall.  Synthesis of Controllers from a Range of Controller Architectures , 1994 .

[14]  P. A. Ruetz,et al.  The architectures and design of a 20-MHz real-time DSP chip set , 1989 .

[15]  Robert W. Brodersen.  Anatomy of a Silicon Compiler , 1992 .

[16]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  R. Palmer,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[18]  Carsten Peterson,et al.  Assessing cereal grain quality with a fully automated instrument using artificial neural network processing of digitized color video images , 1995, Other Conferences.

[19]  Lars Sundström,et al.  Chip for linearisation of RF power amplifiers using digital predistortion , 1994 .

[20]  William K. Pratt,et al.  Digital image processing (2nd ed.) , 1991 .

[21]  Mats Torkelson,et al.  A basic CAD-tool for module generation , 1989, ESSCIRC '89: Proceedings of the 15th European Solid-State Circuits Conference.

[22]  F. Jutand,et al.  A new VLSI architecture for large kernel real time convolution , 1990, International Conference on Acoustics, Speech, and Signal Processing.