Parallel Histogram Calculation for FPGA: Histogram Calculation

Gautam, K. S.

doi:10.1109/iacc.2016.148

Cited by 6 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other architectures with multiple inputs do not mention a specific frequency, so we consider them architecturally. The architecture in [ 26 ] is difficult to run at very high frequency due to the need to connect modules at a long distance, the high fan-out degree, and the complex wiring. The architecture in [ 27 , 28 ] does not really support multiple inputs and outputs at the same time.…”

Section: Methodsmentioning

confidence: 99%

“…For the calculation of HE with multiple inputs and outputs, both of the above architectures have their corresponding variants. Multi-input and -output architectures, such as the multi-decoder architecture, are the first kind [ 26 ]. In this artchitecture, multiple pixels are input to multiple decoders, and the output of the same bit from different decoders is input to the same computing unit.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

An Image Histogram Equalization Acceleration Method for Field-Programmable Gate Arrays Based on a Two-Dimensional Configurable Pipeline

Wang,

Liu,

et al. 2024

Sensors

View full text Add to dashboard Cite

New artificial intelligence scenarios, such as high-precision online industrial detection, unmanned driving, etc., are constantly emerging and have resulted in an increasing demand for real-time image processing with high frame rates and low power consumption. Histogram equalization (HE) is a very effective and commonly used image preprocessing algorithm designed to improve the quality of image processing results. However, most existing HE acceleration methods, whether run on general-purpose CPUs or dedicated embedded systems, require further improvement in their frame rate to meet the needs of more complex scenarios. In this paper, we propose an HE acceleration method for FPGAs based on a two-dimensional configurable pipeline architecture. We first optimize the parallelizability of HE with a fully configurable two-dimensional pipeline architecture according to the principle of adapting the algorithm to the hardware, where one dimension can compute the cumulative histogram in parallel and the other dimension can process multiple inputs simultaneously. This optimization also helps in the construction of a simple architecture that achieves a higher frequency when implementing HE on FPGAs, which consist of configurable input units, calculation units, and output units. Finally, we optimize the pipeline and critical path of the calculation units. In the experiments, we deploy the optimized HE on a VCU118 test board and achieve a maximum frequency of 891 MHz (which is up to 22.6 times more acceleration than CPU implementations), as well as a frame rate of 1899 frames per second for 1080p images.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

An Image Histogram Equalization Acceleration Method for Field-Programmable Gate Arrays Based on a Two-Dimensional Configurable Pipeline

Wang,

Liu,

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

Real Time Catalog of Uniqueness Face Using the CAMShift and Gabor Wavelet Algorithims

Sinaga

Sitio

2019

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Moving image processing uses continuous moving images such as recording live video as a data set. The process of developing a unique database catalog application of student face recognition begins with storing the results of real time recordings of each face. Face tracking uses the CamShift algorithm to identify skin color based on the Hue histogram value. To bring out the special features of a face image that has been convoluted to the kernel, the Gabor Wavelet method is used. 2D Gabor Wavelet kernel which is obtained by modulating 2D sine waves at certain frequencies and orientations with a gaussian envelope. Training data taken from the live video of student activities consists of 30 data sets. Stages start from file conversion - tracking video to multiple frames - RGB-HSV conversion - Calculate Histogram - Face detection and finally face recognition process. If there are similarities with the largest value between querying the face image in the database, the face will be detected. Looping is done to take the face of other students. Video .mp4 is converted to .avi. The results of video conversion are tracked into several frames according to facial movements followed by RGB - HSV conversion, 1-2 minute duration videos produce 40-250 frames. Face recognition successfully identifies moving faces around 80%-90% with skin color detection and 80%-85% with points position detection.

show abstract