Xianfu Xu
ABSTRACT
The 2D Labeling algorithm is used in many applications due to its superior image processing quality. As the requirements of image processing continue to increase, generating huge computational workloads, the efficient real-time implementation of this algorithm is very challenging. In recent years, research on accelerating 2D Labeling algorithms on GPUs has made rapid progress. However, GPU devices usually bring a large amount of energy consumption and are therefore not suitable for a wide range of applications in embedded scenarios. In this paper, we propose a highly integrated general-purpose hardware accelerator for medical image processing to effectively improve the computational performance of 2D Labeling algorithms and reduce the power consumption of FPGA devices. The design integrates image denoising, edge detection, and image segmentation algorithms in a hardware IP core based on a deep pipelining framework, which can effectively improve the speed of 2D Labeling algorithm during intensive medical image processing through parallel computing and data reuse. The design is implemented on Xilinx ZYNQ XC7Z020, and we consume very less energy and improve the computational performance by 1.3 and 2.1 times, respectively, compared to the software design based on advanced NVIDIA GeForce GTX 1660 Super and Intel(R) Core (TM) i7-10700 CPUs.
- Mekali, V. , & Girijamma, H. A. 2019. An Fully Automated CAD System for Juxta-Vacular Nodules Segmentation in CT Scan Images. 2019 3rd International Conference on Computing Methodologies and Communication (ICCMC). IEEE.Google Scholar
- Cao, W. , Wu, R. , Cao, G. , & He, Z. 2020. A comprehensive review of computer-aided diagnosis of pulmonary nodules based on computed tomography scans. IEEE Access, PP(99), 1-1.Google Scholar
- Chen, L. , Bentley, P. , Mori, K. , Misawa, K. , Fujiwara, M. , & D Rueckert. 2018. Drinet for medical image segmentation. IEEE Transactions on Medical Imaging, 1-1.Google Scholar
- Prabukumar, M. , Agilandeeswari, L. , & Ganesan, K. 2019. An intelligent lung cancer diagnosis system using cuckoo search optimization and support vector machine classifier. Journal of ambient intelligence and humanized computing, 10(1), 267-293.Google ScholarCross Ref
- Zhang, C. , Shen, Y. , Kong, Q. , Wei, Y. , Zhang, B. , & Duan, C. , 2019. A novel algorithm for segmentation of solitary pulmonary nodules in chest computed tomography based on three-dimensional connected voxels. Journal of Medical Imaging and Health Informatics.Google ScholarCross Ref
- Enokiya, Y. , Iwamoto, Y. , Chen, Y. W. , & Han, X. H. 2018. Automatic liver segmentation using u-net with wasserstein gans. Journal of Image and Graphics, 6(2), 152-159.Google ScholarCross Ref
- Wang, G. , Li, W. , Zuluaga, M. A. , Pratt, R. , Patel, P. A. , & Aertsen, M. , 2017. Interactive medical image segmentation using deep learning with image-specific fine-tuning. IEEE Transactions on Medical Imaging, PP(99).Google Scholar
- Wang, C., Zhang, J., Li, X., Wang, A., & Zhou, X. 2016. Hardware Implementation on FPGA for Task-Level Parallel Dataflow Execution Engine. IEEE Transactions on Parallel and Distributed Systems, 27(8), 2303-2315. doi:10.1109/TPDS.2015.2487346Google ScholarDigital Library
- Anumala, U. , & Okade, M. 2020. Forensic detection of Median filtering in Images using Local Tetra Patterns and J-Divergence. 2020 National Conference on Communications (NCC).Google ScholarCross Ref
- George, G. , Oommen, R. M. , Shelly, S. , Philipose, S. S. , & Varghese, A. M. 2018. A Survey on Various Median Filtering Techniques For Removal of Impulse Noise From Digital Image. 2018 Conference on Emerging Devices and Smart Systems (ICEDSS). IEEE.Google Scholar
- Wang, X. , Gong, S. , & Liu, J. 2019. Application of improved sobel operator in static gesture segmentation. Electric Engineering.Google Scholar
- Menaka, R. , Janarthanan, S. , & Deeba, K. 2020. Fpga implementation of low power and high speed image edge detection algorithm. Microprocessors and Microsystems, 75, 103053.Google ScholarDigital Library
- Ma, Y. , Ma, H. , & Chu, P. 2020. Demonstration of quantum image edge extration enhancement through improved sobel operator. IEEE Access, 8, 210277-210285.Google ScholarCross Ref
- Ravivarma, G. , Gavaskar, K. , Malathi, D. , Asha, K. G. , & Aarthi, S. 2020. Implementation of sobel operator based image edge detection on fpga. Materials Today: Proceedings(8).Google Scholar
- Zhou Bo, Qin Ling, &Gong Wei. 2019. Stereo-matching algorithm using weighted guided image filtering based on laplacian of gaussian operator. Laser & Optoelectronics Progress, 56(10), 101502.Google ScholarCross Ref
- Ando, T. , & Hiai, F. 2011. Operator log-convex functions and operator means. Mathematische Annalen, 350(3), 611-630.Google ScholarCross Ref
- Huber, M. F. , & Hanebeck, U. D. 2008. Gaussian filter based on deterministic sampling for high quality nonlinear estimation. IFAC Proceedings Volumes, 41(2), 13527-13532.Google ScholarCross Ref
- Benjamini, D. , & Basser, P. J. 2018. Water mobility spectral imaging of the spinal cord: parametrization of model-free laplace mri. Magnetic Resonance Imaging.Google Scholar
- Anshad, P. , Kumar, S. S. , & Shahudheen, S. 2019. Segmentation of chondroblastoma from medical images using modified region growing algorithm. Cluster Computing.Google Scholar
Index Terms
- An FPGA-Based Hardware Accelerator for 2D Labeling
Recommendations
A GPU-Outperforming FPGA Accelerator Architecture for Binary Convolutional Neural Networks
Special Issue on Frontiers of Hardware and Algorithms for On-chip Learning, Special Issue on Silicon Photonics and Regular PapersFPGA-based hardware accelerators for convolutional neural networks (CNNs) have received attention due to their higher energy efficiency than GPUs. However, it is challenging for FPGA-based solutions to achieve a higher throughput than GPU counterparts. ...
A Low-Power Deconvolutional Accelerator for Convolutional Neural Network Based Segmentation on FPGA: Abstract Only
FPGA '18: Proceedings of the 2018 ACM/SIGDA International Symposium on Field-Programmable Gate ArraysConvolutional Neural Networks (CNNs) based algorithms have been successful in solving image recognition problems, showing very large accuracy improvement. In recent years, deconvolution layers are widely used as key components in the state-of-the-art ...
Hardware accelerated FPGA placement
A key advantage of field-programmable gate arrays (FPGAs) over full-custom and semi-custom devices is that they provide relatively quick implementation from concept to physical realization. However, as modern FPGAs reach close to one million logic ...
Comments