Approximate Multiplier for Image Masking Based on Image Multiplication

Abstract

This paper introduces a Verilog Hardware Description Language (HDL) implementation of a specialized multiplier tailored for image processing, with a specific emphasis on edge detection applications. The design integrates an approximate 4-2 compressor and an error correction module, optimizing accuracy while minimizing hardware complexity. Rigorous evaluations, employing ModelSim and MATLAB simulations, validate the design\'s superior accuracy compared to conventional multipliers, achieved with minimal additional hardware complexity. The Verilog HDL implementation is precisely engineered for seamless integration into existing image processing systems, showcasing adaptability and potential across diverse applications. Executed with precision using ModelSim 6.4c and synthesized with the Xilinx Tool, the multiplier strategically incorporates compressors to enhance performance, particularly focusing on efficiency improvement in image multiplication for masking applications. This work strategically converges hardware innovation and image processing intricacies, delving into multiplier design within the realm of edge detection. The proposed system provides an efficient solution, ensuring heightened accuracy and performance while maintaining ease of integration into existing systems. Sobel Image Masking, a fundamental technique in edge detection, further augments the paper\'s practical implications. The multiplication-based approach in Sobel Image Masking employs various advanced multipliers like Booth, Dadda, Shift and add Multipliers, and Wallace multipliers, elucidating their pivotal roles in augmenting gradient values through convolution operations. In alignment with Artificial Intelligence (AI) and Edge Computing, this paper underscores the significance of computational efficiency in image processing. Leveraging Verilog HDL, integrating error correction mechanisms, and applying the specialized multiplier to image masking collectively contribute to advancing AI technologies, especially in edge computing scenarios where task-specific hardware optimization is crucial. Positioned at the intersection of hardware design, image processing, and AI, this paper underscores its relevance in the dynamic landscape of intelligent systems and edge computing, emphasizing the synergy between hardware advancement and artificial intelligence in the realm of edge computing.

Introduction

Multipliers are essential components in DSP and embedded systems, significantly affecting processor speed. High-speed and low-power multiplier designs are crucial due to their heavy use in signal processing tasks like convolution and filtering. Traditional multipliers face challenges such as large area usage, high power consumption, and latency. Modern fast multipliers typically use Booth encoding for partial product generation, carry-save adder trees for partial product reduction, and fast carry-propagate adders for final summation. Compressors like (4:2), (5:2), and (6:2) are popular for improving performance.

For image processing, especially edge detection, existing multipliers often trade off between accuracy and hardware complexity. High precision designs may be accurate but resource-heavy, while approximate or truncated designs reduce complexity but can increase errors. To address this, the proposed multiplier integrates an approximate 4-2 compressor with an error correction module, improving accuracy while maintaining low hardware complexity. Implemented in Verilog HDL, it targets efficient and accurate multiplication for image processing systems.

In edge detection, Sobel Image Masking is widely used, applying horizontal and vertical gradient masks to detect edges by convolving the image with specific kernels. The process involves converting RGB images to grayscale and binary forms, then performing masking to highlight edges, useful for applications like machine vision.

Sobel edge detection utilizes 3×3 convolution masks that smooth noise and emphasize edges, making it preferable over simpler operators like Roberts and Prewitt. The algorithm calculates gradients along the x and y directions, combining them to identify edges based on intensity changes. Additionally, second-order derivative operators like the Laplacian of Gaussian capture rapid intensity changes to enhance edge detection.

The integration of the proposed multiplier with Sobel edge detection offers an efficient solution for real-time image processing, balancing accuracy, hardware efficiency, and performance.

Conclusion

In conclusion, this study introduces an innovative multiplier design tailored for image processing, specifically in the realm of edge detection. The proposed design integrates a novel approximate 4-2 compressor and an error correction module to elevate accuracy while maintaining minimal hardware complexity. Implementation in Verilog HDL and validation through ModelSim and MATLAB simulations underscore the superior accuracy of the proposed multiplier compared to conventional counterparts, with a marginal increase in hardware complexity. The design presents a valuable solution to the challenge of achieving precision and efficiency in multipliers for image processing, particularly in the context of edge detection. Its seamless integration into existing image processing systems holds the promise of improved performance and accuracy. This work signifies a noteworthy advancement in the field of image processing, offering potential enhancements for various applications. To validate our contributions, image masking experiments were conducted utilizing the proposed multiplier. As for future work, there is a scope for further exploration and refinement of the proposed multiplier design. Investigating its performance across diverse datasets and benchmarking against alternative multiplier architectures could provide a comprehensive understanding of its applicability. Additionally, exploring adaptive mechanisms to dynamically adjust precision based on specific image characteristics could enhance the versatility of the multiplier in different image processing scenarios. Furthermore, integrating the multiplier into real world applications and assessing its performance in practical settings would contribute valuable insights. These avenues of exploration could contribute to the ongoing evolution of accurate and efficient multipliers in image processing applications.

References

[1] Parhami, B., & Maghari, N. (2022). Compressor based hybrid approximate multiplier architectures with efficient error correction logic. Microprocessors and Microsystems, 113, 107354. doi:10.1016/j.micpro.2022.107354 [2] Mahdiani, H., Ahmadi, A., Fakhraie, S. M., & Lucas, C. (2023). Design methodology for highly accurate approximate multipliers for error resilient applications. IEEE Transactions on Circuits and Systems I: Regular Papers, 70(1), 492-503. doi:10.1109/TCSI.2022.3196536 [3] J. Liang, J. Han, and F. Lombardi, “New metrics for the reliability of approximate and Probabilistic Adders,” IEEE Trans. Computers, vol. 63, no. 9, pp. 1760–1771, Sep. 2013. [4] V. Gupta, D. Mohapatra, S. P. Park, A. Raghunathan, and K. Roy, “IMPACT: IMPrecise adders for low-power approximate computing,” in Proc. Int. Symp. Low Power Electron. Design, Aug. 2011, pp. 409–414. [5] Shaaban, A. A., Ismail, M., & Abdullah, M. A. (2019). Real-time Sobel edge detection on FPGA for traffic sign recognition. IEEE Access, 7, 122380-122396. [6] Gonzalez, R. C., & Woods, R. E. (2017). Digital image processing. Pearson Education, Inc. (Chapter 4: Edge detection) [7] S. Cheemalavagu, P. Korkmaz, K. V. Palem, B. E. S. Akgul, and L. N. Chakrapani, “A probabilistic CMOS switch and its realization by exploiting [8] noise,” presented at the IFIP Int. Conf. Very Large Scale Integ., Perth, Australia, Oct. 2005. [9] H. R. Mahdiani, A. Ahmadi, S. M. Fakhraie, and C. Lucas, “Bioinspired imprecise computational blocks for efficient VLSI implementation of soft- computing applications,” IEEE Trans. Circuits Syst. I: Reg. Papers, vol. 57, no. 4, pp. 850–862, Apr. 2010. [10] M. J. Schulte and E. E. Swartzlander Jr., “Truncated multiplication with correction constant,” in Proc. Workshop VLSI Signal Process. VI, 1993, pp. 388– 396. [11] E. J. King and E. E. Swartzlander Jr., “Data dependent truncated scheme for parallel multiplication,” in Proc. 31st Asilomar Conf. Signals, Circuits Syst., 1998, pp. 1178–1182. [12] P. Kulkarni, P. Gupta, and M. D. Ercegovac, “Trading accuracy for power in a multiplier architecture,” J. Low Power Electron., vol. 7, no. 4, pp. 490–501, 2011. [13] C. Chang, J. Gu, and M. Zhang, “Ultra low-voltage low- power CMOS 4-2 and 5-2 compressors for fast arithmetic circuits,” IEEE Trans. Circuits Syst., vol. 51, no. 10, pp. 1985–1997, Oct. 2004. [14] D. Radhakrishnan and A. P. Preethy, “Low-Power CMOS pass logic 4-2 compressor for high-speed multiplication,” in Proc. IEEE 43rd Midwest Symp. Circuits Syst., 2000, vol. 3, pp. 1296–1298. [15] Z. Wang, G. A. Jullien, and W. C. Miller, “A new design technique for column compression multipliers,” IEEE Trans. Comput., vol. 44, no. 8, pp. 962– 970, Aug. 1995. [16] J. Gu and C. H. Chang, “Ultra low-voltage, low-power 4-2 compressor for high speed multiplications,” in Proc. 36th IEEE Int. Symp. Circuits Syst., Bangkok, Thailand, May 2003,pp. v-321–v-324. [17] M. Margala and N. G. Durdle, “Low-power low-voltage 4-2 compressors for VLSI Applications,” in Proc. IEEE Alessandro Volta Memorial Workshop Low- Power Design, 1999, pp. 84–90. [18] B. Parhami, Computer Arithmetic; Algorithms and Hardware Designs, 2nd ed. London, U.K.: Oxford Univ. Press, 2010. [19] K. Prasad and K. K. Parhi, “Low-power 4-2 and 5-2 compressors,” in Proc. 35th Asilomar Conf. Signals, Syst. Comput., 2001, vol. 1, pp. 129–133. [20] M. D. Ercegovac and T. Lang, Digital Arithmetic. Amsterdam, The Netherlands: Elsevier, 2003. [21] D. Baran, M. Aktan, and V. G. Oklobdzija, “Energy efficient implementation of parallel CMOS multipliers with improved compressors,” in Proc. ACM/IEEE 16th Int. Symp. Low Power Electron. Design, 2010, pp. 147–152. [22] D. Kelly, B. Phillips, and S. Al-Sarawi, “Approximate signed binary integer multipliers for arithmetic data value speculation,” in Proc. Conf. Design Architect. Signal Image Process., 2009, pp. 97–104. [23] J. Ma, K. Man, T. Krilavicius, S. Guan, and T. Jeong, “Implementation of high performance multipliers based on approximate compressor design,” presented at the Int. SConf. Electrical and Control Technologies, Kaunas, Lithuania, 2011.

Copyright

Copyright © 2025 Rajarajeshwari G, Karthigai Lakshmi S. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.