Motion estimation using maximum sub-image and sub-pixel phase correlation on a DSP platform
- 73 Downloads
Motion estimation is a key step for many video process systems, but it usually suffers from low precision in complex imaging scenarios, as well as the problem of real-time processing. This paper proposes an accurate and fast motion estimation algorithm using maximum sub-image and sub-pixel phase correlation implemented on a DSP platform. Firstly, a maximum sub-image is extracted to contain the image content as much as possible, which both satisfies the requirement of Fourier precondition and the wide coverage of complete background. And then, the extracted sub-image is down-sampled with median filter to increase the signal-noise-ratio and decrease the computation load. Secondly, a sub-pixel motion estimation is used to compensate the losing precision due to down-sampling, and keep the range of motion estimation. Finally, the proposed motion estimation algorithm is implemented on a single core TMS320C6678 DSP platform, and it is accelerated by applying multistage data cache and advanced data access. Experiments demonstrate the accuracy of the proposed motion estimation algorithm in complex scenarios. Meanwhile, it can achieve 7.4 ms/frame for sub-image with 512 × 512 pixels size and 29.6 ms/frame for sub-image with 1024 × 1024 pixels size, respectively.
KeywordsMotion estimation Phase correlation Maximum sub-image Sub-pixel DSP platform
This work is supported by the National Key Research and Development Plan (No.2016YFC0801002), the NSFC (No.61876014, No.61632001, No.61772054) and Army Equipment Research Project (No.301020203).
- 7.Douini Y, Riffi J, Mahraz MA, Tairi H (2017) Solving sub-pixel image registration problems using phase correlation and Lucas-Kanade optical flow method[C]. Intelligent Systems and Computer Vision. p 1–5Google Scholar
- 10.Firouzi S, Joslin C (2013) Motion estimation in blurred frames using phase correlation. International Conference on Signal-Image Technology & Internet-Based Systems. p 201–206Google Scholar
- 12.Ismail Y, McNeely J, Shaaban M, Al Najjar M, Bayoumi MA (2010) A fast discrete transform architecture for frequency domain motion estimation. IEEE International Conference on Image Processing, p 1249–1252Google Scholar
- 18.Matsuo K, Hamada T, Miyoshi M, et al (2009) Accelerating phase correlation functions using GPU and FPGA. NASA/ESA Conference on Adaptive Hardware and Systems, San Francisco, p 433–438Google Scholar
- 19.Monteiro E, Vizzotto B, Diniz C, Zatt B, Bampi S (2011) Applying CUDA architecture to accelerate full search block matching algorithm for high performance motion estimation in video encoding. IEEE International Symposium on Computer Architecture and High Performance Computing, Vitoria, p 128–135Google Scholar
- 20.Reddy BS, Chatterji BN (1996) An FFT-based technique for translation, rotation and scale-invariant image registration. IEEE Trans Image Process 5(8):1266–1271Google Scholar
- 22.Rowekamp T, Platzner M, Peters L (1997) Specialized architectures for optical flow computation: a performance comparison of asic, dsp, and multi-dsp. In: Proceedings of the 8th ICSPAT, p 829–833Google Scholar
- 23.Texas Instruments (2010) TMS320C66x DSP CPU and instruction and set reference guide. Texas, USA, Literature Number: SPRUGH7: 1–1–9–6Google Scholar
- 24.Vasileios M, Ansgar S, Ramesh J, Mohan SK (2013) Real-life events in multimedia: detection, representation, retrieval, and applications. Multimed Tools Appl 70(1):1–6Google Scholar